[LangSmith] Fix env wiring + standardize trace metadata for Workflows LLM calls

[github-actions]

Why

Tracing is only useful if it is consistent and queryable. Some paths set LANGSMITH_API_KEY but not LANGCHAIN_API_KEY, and metadata tags are not consistently attached to runs.

Scope

Ensure LangSmith tracing is enabled consistently when LANGSMITH_API_KEY is present, and add standardized metadata (repo, workflow, run id, issue/PR number, step name) to LLM calls.

Non-Goals

• This does not add a dashboard.
• This does not change core prompts or evaluation logic.

Tasks

• Update tools/llm_provider.py to set LANGCHAIN_API_KEY when LANGSMITH_API_KEY is present and ensure LANGCHAIN_TRACING_V2 is enabled.
• Update scripts/langchain/pr_verifier.py to standardize metadata tagging for LLM calls (include repo, run id, issue/pr number, operation at minimum).
• Update scripts/langchain/followup_issue_generator.py to standardize metadata tagging for LLM calls (include repo, run id, issue/pr number, operation at minimum).
• Add a unit test to verify env wiring behavior when LANGSMITH_API_KEY is set/unset.

Acceptance Criteria

• With LANGSMITH_API_KEY set, LANGCHAIN_TRACING_V2 and LANGCHAIN_API_KEY are configured (covered by a unit test).
• At least two key operations emit traces with standardized metadata fields (inspectable in code, e.g., explicit metadata/tags include repo, run id, issue/pr number, operation).
• With no LANGSMITH_API_KEY, tracing remains off and code paths still run.

Implementation Notes

Primary files:

• tools/llm_provider.py
• scripts/langchain/pr_verifier.py
• scripts/langchain/followup_issue_generator.py

---

Synced by workflow run.

[agents-workflows-bot]

🤖 Auto-pilot step $((STEP_COUNT + 1)): Starting issue formatting...

Running formatter inline (not via label trigger).

[stranske-keepalive]

🤖 Auto-pilot step $((STEP_COUNT + 1)): Analyzing issue for improvements...

Running optimizer inline (not via label trigger).

[stranske-keepalive]

Issue Optimization Suggestions

Review the suggestions below. If you want the agent to apply them, add the
agents:apply-suggestions label to this issue.

Task splitting

• None

Blocked tasks

• None

Objective acceptance criteria

• None

Missing sections

• None

Formatting issues

• None

Notes

• This issue is well-structured and agent-compatible. All required sections are present (Why, Scope, Non-Goals, Tasks, Acceptance Criteria, Implementation Notes). Tasks are appropriately scoped and actionable - each specifies a concrete file to modify and the exact changes needed. The acceptance criteria are objective and testable (env variables configured, metadata fields present, code runs without tracing). No tasks are blocked by agent limitations. The formatting is clean with proper use of checkboxes for tasks and acceptance criteria. The issue provides clear context about the problem (inconsistent tracing configuration) and the solution approach (standardize env variable setup and metadata tagging). File paths are explicitly listed, making implementation straightforward. No splitting needed as each task targets a specific file with a clear deliverable.

[stranske-keepalive]

🤖 Auto-pilot step $((STEP_COUNT + 1)): Applying optimization suggestions...

Running apply inline (not via label trigger).

[agents-workflows-bot]

🤖 Auto-pilot step 1: Issue prepared! Assigning to agent...

Adding agent:codex label. The capability check will run automatically.

⏳ Agent will create a PR shortly.

[stranske-automation-bot]

Codex belt dispatcher queued this issue and created branch codex/issue-1215.

[agents-workflows-bot]

🤖 Auto-pilot step 1: Branch ready, waiting for commits

Branch codex/issue-1215 exists but has no commits yet.
✅ Dispatched the Codex belt worker to generate changes.

📊 Stall count: 1/5 (pauses after 5)

⏳ Auto-pilot will check again on next trigger.

[stranske]

Opened PR #1320 via Codex belt worker.