feat(ui): checkpoint a chat turn, fork from a checkpoint, and duplicate a chat

[Charlesthebird]

---

🤖 written by Claude (start)

Checkpoint a turn in a chat, fork from a checkpoint, duplicate a chat, and the agent rail work that came out of using it.

The Snapshots page and the ListCheckpoints change it needs are stacked on top, in #2798.

Checkpoints and forks

• The Checkpoint button beside Send saves the conversation's current turn boundary. It belongs on the composer rather than a message because CreateCheckpoint takes no cutoff — what it saves is always "as it stands now" — and it is disabled during a turn and once the latest boundary is saved.
• A saved boundary is a line under the turn it was taken at, with Fork on it. The line is the promise: everything above it travels into a fork and nothing below it does. Matched by headTaskId against the turn each message carries, so the lines survive a reload.
• Duplicate chat in each rail row's menu is a checkpoint of the conversation plus a fork of it, and the copy opens.
• A fork refetches its own transcript rather than inheriting the page's, so it never shows its source's boundaries as its own.

The agent rail

Rows gained the room and the states they were missing: a menu on every row carrying Chat details and Share chat as well as rename, duplicate and delete; checkbox hover and press states; controls that line up whether or not the list scrolls. The three columns fold as the window narrows — the agent panel at 1440, the application sidebar at antd's xl, the rail at 1040 — and unfold again. The agent panel's drag handle is gone with ResizableAside: its wrapper clipped to a fixed width, so dragging widened an aside inside a box that never grew.

The create buttons on the list pages are title-cased, which is unrelated to the rest and was asked for separately.

Testing

1. ENABLE_MOCK_UI=true yarn dev, open a conversation, and press Checkpoint beside Send. A line appears under the last turn and is still there after a reload.
2. Press Fork on an early line. The fork opens holding what was above it and nothing below.
3. Open a rail row's menu and choose Duplicate chat. The copy opens with the whole transcript.
4. Narrow the window from wide: the agent panel folds, then the sidebar, then the rail.

The reel in the comments shows this against a real cluster: two checkpoints three turns apart, a fork from each, and the earlier fork knowing only what was above its line. Reels about the Snapshots page moved with it to #2798 and are marked outdated here.

---

🤖 written by Claude (end)

[Charlesthebird]

Note

Outdated. Superseded by the checkpoint-and-fork reel further down this PR.

The original comment

Everything this branch adds, against a real cluster: checkpoints taken from the composer and surviving a reload, forks from two different lines, two duplicates that keep their own favourite city, the agent rail's menus and checkboxes, and the three columns folding as the window narrows.

summary-reel.mp4

---

🤖 written by Claude

[Charlesthebird]

Note

Outdated. The Snapshots page moved to #2798.

The original comment

Snapshots, against a real cluster: three checkpoints taken in a chat, the new page listing every one across all conversations, and three deleted together — one request each, sequential, under one toast.

summary-reel.mp4

---

🤖 written by Claude

[Charlesthebird]

Note

Outdated. Superseded — the listing moved to the controller, in #2798.

The original comment

The Snapshots page here reads every checkpoint into the browser and searches, sorts and pages there. That is deliberate — ListCheckpoints requires an agent_instance_id, so there is no way to ask the controller for a caller's checkpoints at all — but it costs one request per conversation.

#2783 is the follow-up: an RPC that does pagination, search and multi-column sort server-side. It needs a proto change, so it is out of scope here.

---

🤖 written by Claude

[Charlesthebird]

Note

Outdated. This reviewed the branch before the Snapshots page and its listing moved to #2798. The findings still hold, but half of them are against files this PR no longer touches.

The original review

This automated review is a first pass. A manual review will follow once the blockers are addressed.

---

I wrote this branch, so this is a review of my own work: everything below was found, fixed and pushed before posting. Nothing here is outstanding.

The one thing a reviewer should weigh rather than take from me is the Snapshots page reading every checkpoint into the browser. It is a request per conversation, it is deliberate, and the reasoning is in the description and in #2783. If that trade is wrong, the page should not merge in this shape.

What I fixed after review (12)

Level	Where	What was wrong
🔴 High	ChatComposer.tsx	Checkpoint was gated on the streaming phase alone, so it was live between Send and the first token — where CreateCheckpoint refuses with FailedPrecondition. Gated on the whole active turn.
🟠 Medium	ChatTranscript.tsx	Each group was a fragment keyed by its first message. The grouping changes when the boundaries land, so two groups becoming one retired a key and React remounted the messages under it — every rendered mermaid diagram flashed back to loading. One flat run keyed by message id.
🟠 Medium	chat.spec.ts	I had converted a strict assertion into a 20s poll and blamed the checkpoint read. The real cause was the remount above; the poll was hiding it. Strict again.
🟠 Medium	AgentChatPage.tsx	A checkpoint that saved but whose list re-read failed reported Could not checkpoint — over a line already on screen. The re-read is outside the try.
🟠 Medium	grpc/operations.ts	ListCheckpoints read one page and dropped nextPageToken; the controller defaults to 50. Past that the oldest boundaries silently stopped drawing lines. It follows the token.
🟠 Medium	CheckpointDivider.tsx	role="separator" puts its children on ARIA's presentational list, which pruned the label and the only fork control in the application. A labelled group instead.
🟠 Medium	mocks/state.ts	Fixture checkpoints lived in a module array while transcripts persist in sessionStorage, so a checkpoint vanished on reload and the messages it marked did not. The fixture contradicted the controller. Persisted.
🟠 Medium	mocks/state.ts	A second seeded boundary shared a headTaskId with the first, so one line was drawn for two and its Fork silently targeted one of them. Different turn.
🟠 Medium	AgentChatPage.tsx	A fork inherited its source's marks: they are keyed by message id and a fork gets copies of those messages under the same ids. Now held against the conversation they belong to. Guarded by a test I watched fail without the fix.
🟡 Low	ChatComposer.tsx	The tooltip sat on a disabled antd button, which fires no pointer events — invisible in the one state where it explains itself.
🟡 Low	domain/checkpoints.ts	canForkFrom was exported and never called. Deleted.
🟡 Low	rail	Focus rings sliced flat on three controls, each by a different clipping ancestor.

✅ What I checked

• Fork truncation, on a cluster. Forking turn 1 of a five-turn chat gives a chat that knows only the first fact; forking turn 4 knows all four. Not inferred from the SQL — driven in the browser and read back from the agent.
• Duplicates diverge. Two copies of one fork, given different favourite cities, each keep their own after two further messages.
• Marks survive a reload. Three lines saved, reloaded, still three, and the button still refuses a fourth at the same boundary.
• A fork inherits nothing. Reproduced the leak first (fork showed 1 line with 0 checkpoints), then fixed it, then reverted the fix to watch the new test fail.
• Deleting releases the snapshot. Create calls CreateTag, Delete calls DeleteTag; nothing else does. Verified against service.go.
• Both themes, every new control. Photographed at rest, hover, press and focus.
• The collapse order. Measured at seven widths in both directions: agent panel 1440, sidebar xl, rail 1040.
• Control alignment under both scrollbar modes, including --disable-features=OverlayScrollbar, which is the only way the misalignment reproduces.
• CI green, including the lint error I had been misreading locally — I was treating "0 errors and 1 warning potentially fixable" as the total.

Worth a reviewer's judgement

• The fan-out. One request per conversation to draw Snapshots. #2783 has the RPC it needs.
• NotFound for a forked snapshot. The controller refuses to delete one a chat was forked from and reports it as not found. The page states the condition; correcting the error is a backend change.
• Toasts sit over the composer. position="bottom-right" puts them on Send and Checkpoint, so a click is intercepted while one shows. Pre-existing chrome, but this branch made it easy to hit.
• Repeated forks are all named X (fork). Distinguishable only by age in the rail.

---

🤖 written by Claude

[Charlesthebird]

Note

Outdated. The Snapshots page moved to #2798.

The original comment

Snapshots page against a real cluster, 31 snapshots over two pages.

snapshots-server-side.mp4

• Search otter narrows 31 to 2, count and rows together.
• Next and back move between 1–25 of 31 and 26–31 of 31.
• Sorted A–Z, page 2 picks up where page 1 stopped.
• A row's own delete button takes the total to 30.

Turning a page, sorting, and a five-letter search are one ListCheckpoints request each. The conversations are read once per visit — two requests for 70 of them — and only decide which names link to a chat that still exists.

---

🤖 written by Claude

[Charlesthebird]

Note

Outdated. The Snapshots page moved to #2798.

The original comment

What the review changed, against a real cluster with 34 snapshots.

review-fixes.mp4

• Searching 50% finds the one row that contains it, not the 11 the old ILIKE matched on the wildcard. Same for user_id.
• State has no sorter: a listing is READY snapshots only, so it ordered every row by the same value.
• ?page=99 lands back on 1–25 of 34, and the stale page number leaves the address bar.

The index that would keep this off a sequential scan is not here — migration files are immutable once merged, so it needs its own. Tracked in #2783.

---

🤖 written by Claude

[Charlesthebird]

Note

Outdated. The Snapshots page moved to #2798.

The original comment

Injection payloads, and the view living in the address bar.

injection-and-url-state.mp4

• '; DROP TABLE agent_instance_checkpoint; --, ' OR '1'='1 and %' OR 1=1 -- each match no snapshot name and run nothing. The 34 rows are still there afterwards.
• ?q=e&sort=conversation&dir=desc&page=2 opens already searched, sorted and on page 2, and survives a reload.
• Clicking the same controls writes that address back.

Every value reaches Postgres as a bind parameter — the statement is one literal with $1–$7 and pgx prepares it, so nothing is interpolated. The sort column is an enum on the wire and is whitelisted against a map before the query runs; the filter is position(lower($3) IN …), so it has no pattern syntax to escape.

---

🤖 written by Claude

[Charlesthebird]

The Snapshots page moved to #2798, stacked on this branch.

This PR is now the chat side only: checkpoint a turn, fork from a checkpoint, duplicate a chat, and the agent rail work. No proto, no Go — 30 files, all under ui/.

#2798 carries the page itself, the ListCheckpoints change it needs to page, search and sort across conversations, and the store and service work behind that. It closes #2783.

Merge this one first; GitHub retargets #2798 to main once it lands.

The reels above are marked outdated for that reason — the ones about snapshots belong with the other PR now.

---

🤖 written by Claude

[Charlesthebird]

Checkpoint and fork against a real cluster, with a real agent answering.

checkpoint-and-fork.mp4

One conversation, two checkpoints three turns apart, then a fork from each:

• After my favorite color is teal, Checkpoint. Two more turns — the otter, Kyoto — then Checkpoint again. Two lines at different depths.
• Fork the first line: two messages, and asked what it knows it says teal. The otter and Kyoto were below the line.
• Fork the second line: six messages, and it lists teal, the otter and Kyoto.
• Fork the first line again: another two-message copy. A boundary is not spent by forking it.

Neither fork shows the source's lines as its own — a fork reads its own boundaries.

---

🤖 written by Claude