Skip to content

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Jose-Gael-Cruz-Lopez merged 2 commits into from
May 5, 2026
Merged

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) #78

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
77ff33f
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
Jose-Gael-Cruz-Lopez May 5, 2026
be5b842
fix(learn): close PR #78 review issues — model symmetry + declared se…
Jose-Gael-Cruz-Lopez May 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions backend/agents/_providers.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
# call where the agent pulls structured graph data via tools, so the
# bulk of the value is in tool wiring, not raw model strength.
"quiz": "gemini-2.5-flash-lite",
# Chat tutor runs on Pro: it streams a multi-turn pedagogical
# conversation where reasoning depth and instruction following drive
# perceived quality. Matches main's tutor default after PR #73
# (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
# (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
"chat_tutor": "gemini-2.5-pro",
}


Expand Down
134 changes: 134 additions & 0 deletions backend/agents/chat_tutor.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
read_session_history_tool,
read_user_progress_tool,
search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
"You are Sapling, an AI tutor that helps a student build mastery in "
"their course material. You have tools to fetch the student's "
"progress, search their uploaded course documents, and update their "
"knowledge graph mastery scores. Use tools when relevant — don't "
"fabricate context.\n\n"
"Tone: warm, concise, no filler. Use math/code blocks where helpful "
"(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
"MODE: Socratic. Lead the student to the answer through questions, "
"not lectures. Each turn: ask one focused question that reveals what "
"they already know or where they're confused. Avoid giving the answer "
"directly; provide hints only after they've made an attempt. End "
"every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
"MODE: Expository. Explain the concept directly and thoroughly. "
"Structure your response: brief overview → detailed explanation → "
"concrete example or worked problem. Don't ask questions back unless "
"the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
"MODE: TeachBack. The student is teaching you a concept. Listen to "
"their explanation, then identify what's correct, what's missing, "
"and any specific misconceptions. Praise accuracy where it exists. "
"End with one targeted question that probes the weakest spot in "
"their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
"socratic": _SOCRATIC_PROMPT,
"expository": _EXPOSITORY_PROMPT,
"teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
search_course_materials_tool,
read_session_history_tool,
read_user_progress_tool,
apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
return Agent[SaplingDeps, str](
model=model_for("chat_tutor"),
deps_type=SaplingDeps,
output_type=str,
system_prompt=_PROMPTS[mode],
metadata={
"prompt_version": _PROMPT_HASHES[mode],
"agent": "chat_tutor",
"mode": mode,
},
tools=_TOOLS,
)


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
"""Return the agent instance for a given mode string.

Falls back to Socratic if the mode is unrecognized (or missing) —
same default the legacy `build_system_prompt` used when no mode
matched the MODE_PROMPTS dict.
"""
normalized = (mode or "socratic").lower()
return {
"socratic": socratic_agent,
"expository": expository_agent,
"teachback": teachback_agent,
}.get(normalized, socratic_agent)
5 changes: 5 additions & 0 deletions backend/agents/deps.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
version.
request_id: A correlation ID for tracing across a single
user-facing request. Used by Logfire spans.
session_id: The active chat session, when applicable. Used by tools
that need to scope reads to *this* conversation (e.g.
read_session_history_tool). Optional — agent runs that don't
happen inside a session (eval mode, batch tasks) leave it None.
"""

user_id: str
course_id: str | None
supabase: Any
request_id: str
session_id: str | None = None
Loading
Loading
, 'i'); if (__m === '*' || __re.test(location.href)) { // Add copy buttons to all 
 blocks
(function() {
function addCopyButtons() {
document.querySelectorAll('pre code').forEach(function(codeBlock) {
if (codeBlock.parentElement.hasAttribute('data-copy-added')) return;
codeBlock.parentElement.setAttribute('data-copy-added', 'true');
var btn = document.createElement('button');
btn.textContent = 'Copy';
btn.style.cssText = 'position:absolute;top:4px;right:4px;padding:2px 8px;font-size:11px;background:#4ecdc4;border:none;border-radius:4px;color:#1a1a2e;cursor:pointer;opacity:0.7;transition:opacity 0.2s;';
btn.onmouseover = function() { this.style.opacity = '1'; };
btn.onmouseout = function() { this.style.opacity = '0.7'; };
btn.onclick = function() {
navigator.clipboard.writeText(codeBlock.textContent).then(function() {
btn.textContent = 'Copied!';
setTimeout(function() { btn.textContent = 'Copy'; }, 1500);
});
};
codeBlock.parentElement.style.position = 'relative';
codeBlock.parentElement.appendChild(btn);
});
}
addCopyButtons();
// Re-run on dynamic content
var observer = new MutationObserver(addCopyButtons);
observer.observe(document.body, { childList: true, subtree: true });
})();
}
} catch(__e) { console.warn('[Userscript:Add Copy Buttons to Code Blocks]', __e); }
})();
(function(){
try {
var __m = "github.com";
var __re = new RegExp('^' + "github\\.com" + '
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading

, 'i');
if (__m === '*' || __re.test(location.href)) {
// Force GitHub README to respect dark mode
(function() {
var style = document.createElement('style');
style.textContent = '
.markdown-body {
color-scheme: dark light;
}
.markdown-body pre { background: #161b22 !important; }
.markdown-body code { background: rgba(110, 118, 129, 0.4) !important; }
.markdown-body table th, .markdown-body table td { border-color: #30363d !important; }
.markdown-body img { background: #0d1117; }
.markdown-body blockquote { border-left-color: #8b949e; }
.markdown-body hr { border-color: #30363d; }
';
document.head.appendChild(style);
})();
}
} catch(__e) { console.warn('[Userscript:GitHub Dark Mode README Fix]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading

, 'i');
if (__m === '*' || __re.test(location.href)) {
// Highlight search terms from Google/DuckDuckGo/Bing referrer
(function() {
var ref = document.referrer;
var terms = [];
if (ref.includes('google.com') || ref.includes('duckduckgo.com') || ref.includes('bing.com')) {
var url = new URL(ref);
var q = url.searchParams.get('q') || url.searchParams.get('p');
if (q) {
terms = q.split(/\s+/).filter(function(t) { return t.length > 2; });
}
}
if (terms.length === 0) return;
var style = document.createElement('style');
style.textContent = '.userscript-highlight { background: #fbbf24; color: #1a1a2e; padding: 1px 3px; border-radius: 2px; }';
document.head.appendChild(style);
function highlight(node) {
if (node.nodeType === 3) { // text node
var text = node.textContent;
var found = false;
terms.forEach(function(term) {
var regex = new RegExp('(' + term.replace(/[.*+?^${}()|[\]\\]/g, '\\') + ')', 'gi');
if (regex.test(text)) {
found = true;
var frag = document.createDocumentFragment();
var parts = text.split(regex);
parts.forEach(function(part, i) {
if (i % 2 === 0) {
frag.appendChild(document.createTextNode(part));
} else {
var span = document.createElement('span');
span.className = 'userscript-highlight';
span.textContent = part;
frag.appendChild(span);
}
});
node.parentNode.replaceChild(frag, node);
}
});
} else if (node.nodeType === 1 && node.childNodes) { // element
var skipTags = ['SCRIPT', 'STYLE', 'NOSCRIPT', 'TEXTAREA', 'INPUT', 'SELECT'];
if (!skipTags.includes(node.tagName)) {
Array.from(node.childNodes).forEach(highlight);
}
}
}
highlight(document.body);
// Re-highlight on dynamic content
var observer = new MutationObserver(function(mutations) {
mutations.forEach(function(m) {
m.addedNodes.forEach(function(node) {
if (node.nodeType === 1 || node.nodeType === 3) highlight(node);
});
});
});
observer.observe(document.body, { childList: true, subtree: true });
})();
}
} catch(__e) { console.warn('[Userscript:Highlight Search Terms]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading

, 'i');
if (__m === '*' || __re.test(location.href)) {
// Strip utm_, fbclid, gclid, etc. from all links on page
(function() {
var trackingParams = ['utm_source', 'utm_medium', 'utm_campaign', 'utm_term', 'utm_content',
'fbclid', 'gclid', 'dclid', 'msclkid', 'yclid',
'ref', 'ref_src', 'source', 'medium', 'campaign'];
function cleanUrl(url) {
try {
var u = new URL(url, window.location.origin);
var changed = false;
trackingParams.forEach(function(p) {
if (u.searchParams.has(p)) {
u.searchParams.delete(p);
changed = true;
}
});
return changed ? u.toString() : url;
} catch (e) {
return url;
}
}
function cleanLinks() {
document.querySelectorAll('a[href]').forEach(function(a) {
var clean = cleanUrl(a.href);
if (clean !== a.href) a.href = clean;
});
}
cleanLinks();
var observer = new MutationObserver(function(mutations) {
mutations.forEach(function(m) {
m.addedNodes.forEach(function(node) {
if (node.nodeType === 1) {
if (node.tagName === 'A') cleanLinks();
node.querySelectorAll('a[href]').forEach(function(a) {
var clean = cleanUrl(a.href);
if (clean !== a.href) a.href = clean;
});
}
});
});
});
observer.observe(document.body, { childList: true, subtree: true });
})();
}
} catch(__e) { console.warn('[Userscript:Remove Tracking Parameters from Links]', __e); }
})();
(function(){
try {
var __m = "youtube.com";
var __re = new RegExp('^' + "youtube\\.com" + '
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading

, 'i');
if (__m === '*' || __re.test(location.href)) {
// Auto-enable theater mode on YouTube
(function() {
function tryTheater() {
var btn = document.querySelector('button[aria-label="Theater mode"], ytd-player #player button[title="Theater mode"]');
if (btn && !btn.classList.contains('activated')) {
btn.click();
}
}
// Try immediately
tryTheater();
// Try after navigation (SPA)
var lastUrl = location.href;
setInterval(function() {
if (location.href !== lastUrl) {
lastUrl = location.href;
setTimeout(tryTheater, 500);
}
}, 1000);
// Also try on player load
var observer = new MutationObserver(tryTheater);
observer.observe(document.body, { childList: true, subtree: true });
})();
}
} catch(__e) { console.warn('[Userscript:YouTube Theater Mode Default]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading

, 'i');
if (__m === '*' || __re.test(location.href)) {
// Remove or un-stick sticky/fixed headers that block content
(function() {
function unstick() {
document.querySelectorAll('header, nav, [role="banner"], .header, .navbar, .sticky, .fixed-top, [style*="position: fixed"], [style*="position:sticky"]').forEach(function(el) {
if (el.style.position === 'fixed' || el.style.position === 'sticky' || getComputedStyle(el).position === 'fixed' || getComputedStyle(el).position === 'sticky') {
el.style.position = 'static';
el.style.top = 'auto';
el.style.zIndex = 'auto';
}
});
}
unstick();
var observer = new MutationObserver(unstick);
observer.observe(document.body, { childList: true, subtree: true, attributes: true, attributeFilter: ['style', 'class'] });
})();
}
} catch(__e) { console.warn('[Userscript:Kill Sticky Headers]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading

, 'i');
if (__m === '*' || __re.test(location.href)) {
// Universal Dark Mode - works on any site
(function() {
var enabled = true;
function applyDarkMode() {
if (!enabled) return;
// Create style element if it doesn't exist
var style = document.getElementById('universal-dark-mode-style');
if (!style) {
style = document.createElement('style');
style.id = 'universal-dark-mode-style';
document.head.appendChild(style);
}
// Dark mode CSS - inverts colors but preserves images/video
style.textContent = '
/* Invert everything except media */
html {
filter: invert(1) hue-rotate(180deg) !important;
background: #1a1a2e !important;
}
/* Restore images, videos, iframes, canvas */
img, video, iframe, canvas, svg, picture, [style*="background-image"] {
filter: invert(1) hue-rotate(180deg) !important;
}
/* Preserve specific elements that should not be inverted */
.no-dark-mode, .no-dark-mode *,
[data-theme="light"], [data-theme="light"],
.ace_editor, .ace_editor *,
.CodeMirror, .CodeMirror *,
.monaco-editor, .monaco-editor *,
.markdown-body pre, .markdown-body pre *,
.highlight, .highlight *,
pre code, pre code * {
filter: none !important;
}
/* Fix common UI elements */
.modal, .popup, .dropdown-menu, .tooltip, .popover {
filter: invert(1) hue-rotate(180deg) !important;
background: #2d2d44 !important;
border-color: #444 !important;
}
/* Scrollbars */
::-webkit-scrollbar { background: #1a1a2e !important; }
::-webkit-scrollbar-thumb { background: #444 !important; }
::-webkit-scrollbar-thumb:hover { background: #555 !important; }
/* Selection */
::selection { background: #4ecdc4 !important; color: #1a1a2e !important; }
::-moz-selection { background: #4ecdc4 !important; color: #1a1a2e !important; }
';
}
function removeDarkMode() {
var style = document.getElementById('universal-dark-mode-style');
if (style) style.remove();
}
// Toggle with Alt+Shift+D
document.addEventListener('keydown', function(e) {
if (e.altKey && e.shiftKey && e.key === 'D') {
e.preventDefault();
enabled = !enabled;
if (enabled) {
applyDarkMode();
console.log('[Universal Dark Mode] Enabled');
} else {
removeDarkMode();
console.log('[Universal Dark Mode] Disabled');
}
}
});
// Apply on load
applyDarkMode();
// Re-apply on dynamic content
var observer = new MutationObserver(function(mutations) {
if (enabled && !document.getElementById('universal-dark-mode-style')) {
applyDarkMode();
}
});
observer.observe(document.head, { childList: true });
console.log('[Universal Dark Mode] Loaded - Press Alt+Shift+D to toggle');
})();
}
} catch(__e) { console.warn('[Userscript:Universal Dark Mode]', __e); }
})();
})();
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3) by Jose-Gael-Cruz-Lopez · Pull Request #78 · SaplingLearn/Sapling · GitHub
Skip to content
refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)#78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Jose-Gael-Cruz-Lopez
merged 2 commits into

from

May 5, 2026
 Merged


refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)
#78
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
77ff33f

refactor(learn): convert chat tutor to chat_tutor_agent (refactor #3)


Jose-Gael-Cruz-Lopez May 5, 2026
be5b842

fix(learn): close PR #78 review issues — model symmetry + declared se…


Jose-Gael-Cruz-Lopez May 5, 2026

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


12 changes: 10 additions & 2 deletions
backend/agents/_providers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -8,9 +8,11 @@
 SAPLING_MODEL_CONCEPTS=gemini-2.5-flash
 SAPLING_MODEL_SYLLABUS=gemini-2.5-flash
 SAPLING_MODEL_QUIZ=gemini-2.5-flash-lite
 SAPLING_MODEL_CHAT_TUTOR=gemini-2.5-pro

Defaults are tuned per task: cheaper models for simpler classifications,
flagship Flash for tasks where output quality drives downstream UX.
flagship Flash for tasks where output quality drives downstream UX, and
the Pro tier for the conversational tutor where reasoning depth matters.
"""

from __future__ import annotations
Expand All@@ -24,7 +26,7 @@
from config import GEMINI_API_KEY


AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz"]
AgentTask = Literal["classifier", "summary", "concepts", "syllabus", "quiz", "chat_tutor"]


# Defaults are conservative. Bumping a model up costs more; the env var
Expand All@@ -38,6 +40,12 @@
 # call where the agent pulls structured graph data via tools, so the
 # bulk of the value is in tool wiring, not raw model strength.
 "quiz": "gemini-2.5-flash-lite",
 # Chat tutor runs on Pro: it streams a multi-turn pedagogical
 # conversation where reasoning depth and instruction following drive
 # perceived quality. Matches main's tutor default after PR #73
 # (`feat(learn): use gemini-2.5-pro for tutor chat`) and PR #74
 # (`fix(learn): allow thinking on gemini-2.5-pro multiturn calls`).
 "chat_tutor": "gemini-2.5-pro",
}


Expand Down

134 changes: 134 additions & 0 deletions
backend/agents/chat_tutor.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
"""Chat tutor agent for the Learn route's three teaching modes.

Replaces routes/learn.py:152's build_system_prompt + call_gemini_multiturn
with a typed Pydantic AI agent. Tools handle the data lookups that used
to be string-stuffed: search_course_materials, read_session_history,
read_user_progress, apply_graph_update_tool.

Modes (Socratic, Expository, TeachBack) are gated by selecting different
system prompts at construction time. The route picks the right agent
instance per request based on body.mode.
"""

from __future__ import annotations

import hashlib
from typing import Literal

from pydantic_ai import Agent

from agents._providers import model_for
from agents.deps import SaplingDeps
from agents.tools.chat_context import (
 read_session_history_tool,
 read_user_progress_tool,
 search_course_materials_tool,
)
from agents.tools.graph import apply_graph_update_tool


TutorMode = Literal["socratic", "expository", "teachback"]


# ── System prompts (one per mode) ──────────────────────────────────────────

# The shared preamble is identical across modes so a prompt-version bump
# in shared guidance shows up as a hash change for every mode at once.
_SHARED_PREAMBLE = (
 "You are Sapling, an AI tutor that helps a student build mastery in "
 "their course material. You have tools to fetch the student's "
 "progress, search their uploaded course documents, and update their "
 "knowledge graph mastery scores. Use tools when relevant — don't "
 "fabricate context.\n\n"
 "Tone: warm, concise, no filler. Use math/code blocks where helpful "
 "(LaTeX `$x^2$`, ```mermaid```, ```plot```). Don't over-explain.\n\n"
)

_SOCRATIC_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Socratic. Lead the student to the answer through questions, "
 "not lectures. Each turn: ask one focused question that reveals what "
 "they already know or where they're confused. Avoid giving the answer "
 "directly; provide hints only after they've made an attempt. End "
 "every response with a question."
)

_EXPOSITORY_PROMPT = _SHARED_PREAMBLE + (
 "MODE: Expository. Explain the concept directly and thoroughly. "
 "Structure your response: brief overview → detailed explanation → "
 "concrete example or worked problem. Don't ask questions back unless "
 "the student's prompt is genuinely ambiguous."
)

_TEACHBACK_PROMPT = _SHARED_PREAMBLE + (
 "MODE: TeachBack. The student is teaching you a concept. Listen to "
 "their explanation, then identify what's correct, what's missing, "
 "and any specific misconceptions. Praise accuracy where it exists. "
 "End with one targeted question that probes the weakest spot in "
 "their understanding."
)

_PROMPTS: dict[TutorMode, str] = {
 "socratic": _SOCRATIC_PROMPT,
 "expository": _EXPOSITORY_PROMPT,
 "teachback": _TEACHBACK_PROMPT,
}

# Hash of each mode's full prompt (preamble + body), for span versioning.
# Logfire spans on chat-tutor runs include this so a prompt revision
# shows up as a clean delta when comparing run metadata across deploys.
_PROMPT_HASHES: dict[TutorMode, str] = {
 mode: hashlib.sha256(prompt.encode("utf-8")).hexdigest()[:12]
 for mode, prompt in _PROMPTS.items()
}


# ── Agent (one per mode, sharing the same tool surface) ────────────────────

# Output type is plain str — chat tutor produces free-form Markdown that
# the frontend renders via MarkdownChat. No structured output here; that
# is reserved for routes that grade or extract.

# All four tools are registered on every mode. The system prompt steers
# WHEN to call them; the surface stays uniform so a Pro-tier model can
# decide for itself which lookups are worth the round trip.
_TOOLS = [
 search_course_materials_tool,
 read_session_history_tool,
 read_user_progress_tool,
 apply_graph_update_tool,
]


def _build_agent(mode: TutorMode) -> Agent[SaplingDeps, str]:
 return Agent[SaplingDeps, str](
 model=model_for("chat_tutor"),
 deps_type=SaplingDeps,
 output_type=str,
 system_prompt=_PROMPTS[mode],
 metadata={
 "prompt_version": _PROMPT_HASHES[mode],
 "agent": "chat_tutor",
 "mode": mode,
 },
 tools=_TOOLS,
 )


socratic_agent = _build_agent("socratic")
expository_agent = _build_agent("expository")
teachback_agent = _build_agent("teachback")


def agent_for_mode(mode: str | None) -> Agent[SaplingDeps, str]:
 """Return the agent instance for a given mode string.

 Falls back to Socratic if the mode is unrecognized (or missing) —
 same default the legacy `build_system_prompt` used when no mode
 matched the MODE_PROMPTS dict.
 """
 normalized = (mode or "socratic").lower()
 return {
 "socratic": socratic_agent,
 "expository": expository_agent,
 "teachback": teachback_agent,
 }.get(normalized, socratic_agent)

5 changes: 5 additions & 0 deletions
backend/agents/deps.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -23,9 +23,14 @@ class SaplingDeps:
 version.
 request_id: A correlation ID for tracing across a single
 user-facing request. Used by Logfire spans.
 session_id: The active chat session, when applicable. Used by tools
 that need to scope reads to *this* conversation (e.g.
 read_session_history_tool). Optional — agent runs that don't
 happen inside a session (eval mode, batch tasks) leave it None.
 """

 user_id: str
 course_id: str | None
 supabase: Any
 request_id: str
 session_id: str | None = None
Loading
Loading