Skip to content

fix(dashboard): show the GitHub PRs a chat owns on its card, not every PR it mentions - #35327

Closed
kyleseaman wants to merge 2 commits into
mainfrom
fix/github-link-detection
Closed

kyleseaman wants to merge 2 commits into
mainfrom
fix/github-link-detection

Conversation

@kyleseaman

@kyleseaman kyleseaman commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: A session card in the Crew sidebar shows the GitHub pull requests and issues the chat actually opened, worked on or reported, including those its workers opened, instead of every GitHub link its messages happen to mention.

A card's GitHub chips came from any PR or issue URL in the text of any row. Two things went wrong:

Why it matters

The chips are how the owner finds the PR a chat is working on from the sidebar. When the right PR is missing and the card is full of PRs the chat only talked about, they have to open the chat and search the transcript, and the three visible chips are often the wrong three.

Not a goal

  • A pin/attach action for links a chat only mentions.
  • Status refresh for chips revealed by the "+N" expand.
  • Detecting a PR the chat only pushed to without any gh write or report.
  • Resolving gh pr merge 5 / gh pr review 5 run with no -R and no URL: gh prints nothing off a terminal, so the row does not say which repository it was.
  • Reading the shell's ambient GH_HOST: with GitHub Enterprise hosts configured, a host-less owner/repo counts only with a GH_HOST= on the call itself.
  • Hiding a work board's PRs on /clear: /clear starts the conversation over, so the chips its rows named go, but the board keeps its items, so the card keeps the PRs they report.
  • Any change to how chips look, or to the "+N" toggle.
  • Changing GitLab, Jira, Gerrit, Bitbucket Server or plugin links. They have no ownership signal, so a mention still makes their chip.
  • Changing URL host or path validation in source_providers/links.py.
  • Known limits of reading compound shell commands, which make no chip or a wrong one (each one needs per-call output, which a tool row does not have):
    • On the create path, a bare PR URL line printed by another program or by a target-less gh read (gh pr list --jq '.[].url') reads as created.
    • Two repositories sharing one bare PR number in one command; a $n loop target credits only the first PR.
    • gh ... || exit 1 and gh ...; echo "exit=$?" count as masked, so a non-TTY write that prints nothing makes no chip.
    • Any subshell ( ... ) in the command makes every gh call in it unproven.
    • A line break inside an unquoted $(...) (for example a multi-line <(cat <<'EOF' ... EOF)) stays in the blanked text and splits the command, so a gh call on that line counts as unproven.
    • A row from before this change, on a backend whose shell output has no Exit Code: trailer, makes no chip: it stored no completed status, so nothing shows it ran.
    • A target-less write next to another gh call whose output names the same PR (gh pr edit --add-label x && gh pr view) makes no chip.

What changed (motivation → approach → change)

The sidebar scanned row text for URLs; the signals that say who owns a PR live in tool output and in the work ledger. A GitHub link now counts for a card only when one of these holds:

Signal Where it is read
Created: a gh pr create / gh issue create row that exited 0 the URL line gh printed
Acted on: a gh pr edit/comment/close/merge/ready/review or a writing gh api .../pulls/<N> that exited 0 the target URL, -R plus number, or gh's own pull request owner/repo#N
Reported: the chat's own work-ledger item its pr (with the acceptance repo) and PR URLs in its artifacts
Rolled up: a worker bound to one of this chat's items what that worker created or reported, while it is still bound to that item; one level, and a worker's acted-on PRs stay on its card

Reads (gh pr view, checks, GET gh api) and --help never count; what writes is decided by the batch guard's own classifier (acp/batch_guard._gh_is_write), and a gh api endpoint is found by skipping the guard's own valued-option table (_GH_VALUED_OPTIONS), both read through agent_sdk, so the card and the guard agree. A row counts only on proof that its call ran and succeeded: a shell Exit Code: 0 trailer, or the call's stored ACP status completed. A refused, user-denied, failed, cancelled or still-running row never counts, whatever its argv names. A write whose exit the row's code does not vouch for (|| true, a pipe after it) counts only when gh's output names the PR. The gh proxy prints gh's own output and exit code, so its rows count the same way. Chips are ordered by the newest owning event, so a compaction summary cannot reorder them. A host-less owner/repo reads as github.com only once the host settings have loaded and no Enterprise host is configured.

Piece Change
source_providers/ownership.py (new), agent_sdk/__init__.py The tool-row reader (shell splitting, exit trailer, created/targeted PR) and the work-ledger reader, which runs on the default executor, keeps its answer until a write to a board it came from, an allowlist change or 60 seconds, and pushes the slots when a read lands changed. agent_sdk hands it the batch guard's gh classifier and tables (gh_call_writes, gh_write_verbs, gh_valued_options, command_positionals), so it keeps no copy and no kiro_crew.acp import.
slot_projection.py, state_parts/source_links.py, slot_directory.py, chat_slot.py GitHub links come from tool rows (other providers still scan text), memoized per row; _pr_source_links merges own, ledger and worker links by identity, newest first, with unlinks applied to all; a conductor looks up its workers' live slots.
work_ledger_store.py, source_providers/hosts.py, chat_turn/tool_results.py, restore_drivers.py, chat_persistence.py What keeps the answer fresh: a bind moves the worker's change mark, the first host-settings load moves the allowlist generation, each tool row stores its call's terminal status (meta.tool_status), a gh-write row's output or status invalidates the chip cache, and a restored sidebar tail keeps a gh-write row's command, output end (redacted, bounded) and status.
website/src/utils/sourceProviderMeta.ts sourceTabQualifier treats KiroCrew and kirocrew as one GitHub project, matching the backend identity (#34650).

The chat view's Changes and Resources panels build their own list in the browser and are unchanged.

Backwards compatibility

Compatible: the slot payload, the expand route and the unlink route keep their shapes and keys. Stored unlink tombstones still match.

Reader: src/kiro_crew/dashboard/state_parts/source_links.py:_pr_source_links -- chat -- a GitHub link only mentioned in text leaves the card; test_a_row_that_only_mentions_a_pr_makes_no_chip
Reader: src/kiro_crew/dashboard/chat_api/source_links.py:api_chat_slot_source_links -- crew page -- the "+N" expand serves the same owned list; test_dashboard_source_links_expand.py
Reader: src/kiro_crew/dashboard/chat_api/source_links.py:_apply_source_link_unlink -- crew page -- derived identities include rolled-up links, and the in-lock re-check reads tool output; test_an_unlinked_rolled_up_chip_stays_hidden, test_an_owned_url_only_in_meta_still_authorizes_its_unlink
Reader: src/kiro_crew/dashboard/state_parts/source_links.py:_slot_chip_urls -- crew page -- status refresh follows the owned chips; test_dashboard_state_ws_status_refresh.py
Reader: website/src/pages/chat/useChatPageResourcesController.tsx -- chat -- the Changes/Resources panels extract links in the browser and do not read this list; unchanged

Tests

test/test_source_link_ownership.py, one test per symptom, each failing on origin/main:

  • A gh pr create row with the URL only in its output gives 1 chip (chat-539's feat(app-sdk): issue list components (useIssues, IssueList, IssueCard) #34732 row).
  • A conductor whose worker reported pr=34732 shows it on the conductor card and on the worker's own (chat-537 / chat-539).
  • User, assistant, nudge and system rows that only mention a PR give no chip.
  • A compaction summary re-mentioning an old PR keeps it behind a newer created PR.
  • A worker report under kirodotdev/kirocrew and a created kirodotdev/KiroCrew URL give 1 chip; website/src/test/sourceTabQualifier.githubCase.test.ts covers the qualifier (its two-spellings case fails on origin/main).

Nearby inputs tested: failed and refused creates (exit 1, "already exists" behind || true), reads and --help (not counted), gh issue create, another repository, a fork PR, every write verb and the review/merge switches, bare numbers resolved from gh's output, a URL in --body, writing gh api (relative, absolute Enterprise, {owner} placeholders), masked writes, heredocs, here-strings, line continuations and 2>&1, direct argv, artifacts URLs and owner/repo#N, a reported number with no repo, a worker that created without reporting or edited after creating, one-level roll-up, newest-event order across sources, unlinked rolled-up chips, a channel-born conductor, a bind after a card read, cold and Enterprise host settings, a restored sidebar tail, /clear (own chips go, board chips stay), the gh proxy output shape, non-ASCII and huge PR numbers, a bounded board answer, cache eviction, the ledger read happening off the loop, a 100-call command over 1 MB of output reading its output tail once, and compound commands: another gh call's or another program's printed URL (bare numbers, -R mixes, {owner}/{repo} api paths, masked writes), pipelines gh starts, if/for/{ } blocks, !, quoted and unquoted $(...) and backticks (also in heredoc prose), and operators that end a line. The TestAdversarialInputs and TestReviewFindings* classes pin what a fresh-context break attempt and the local GPT/Opus review prompts found. Each was shown red with the production change reverted.

test/test_session_card_source_links_knob.py gains an autouse fixture restoring the chip switch: its config-PUT test left "off" in the process-wide snapshot, so later tests in the same xdist worker serialized no chips (4 failures when it runs before test_source_link_github_case.py, none with the fixture).

The existing chip suites built their chips from assistant text; they now build them from a gh pr create row (test/source_link_rows.py). Tests of the text scanner's own mechanics moved to gitlab.com merge-request URLs, which still chip from a mention.

Determinism: test_source_link_ownership.py passes 10 runs at -n0 and 3 shuffled orders; the 74 files the change touches or that import a changed module pass at -n 2.

Proof: not provable -- other: prove.py reports BASELINE_RED in its throwaway worktree on this host; every production hunk was reverted by hand instead and its tests shown red

Manual verification

On copies of real transcripts (no live chat touched): chat-539 now derives #34732 and chat-538 #34742, which neither card showed before. A warm re-derivation of the 4,618-row Slack Code conductor takes 11 ms, against 14 ms on origin/main.

Related Issues

N/A — no tracking issue.

Sibling sweep

Swept: git grep -nE "iter_source_url_candidates|_pr_source_links|source_links(" -- src/kiro_crew/dashboard website/src
Sibling: src/kiro_crew/dashboard/state_parts/source_links.py -- changed here
Sibling: website/src/utils/pullRequestLinks.ts -- not affected -- the chat view's own panel extractor, which the goal leaves as it is
Sibling: website/src/utils/extractChatLinks.ts -- not affected -- chat nav links from user/assistant rows, not card chips

Pattern harvest

Not generalizable: the rule is specific to which GitHub signals mean a chat owns a PR, and it now lives in the one derivation every chip consumer reads.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

@kyleseaman
kyleseaman requested review from a team and KiroCrewCaptain as code owners November 11, 2026 14:22
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Nov 11, 2026
@github-actions

github-actions Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS (a BLOCK with no Trigger and no Outcome does not block)

UX-level review of 948d4729a14048ff32c95fd147782cd41e2e4b13 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

cannot evaluate: missing a screenshot of a session card with the new chip set. This PR changes which GitHub chips render on Crew sidebar cards, and their qualifier text, but supplies no image, so the blind read was not performed.

Evidence gaps

  • Which chips a card shows (SessionSourceChips). Chips for PRs the chat only mentioned now leave the card, and chips for PRs its workers created or reported now appear on the conductor's card. To close: before/after screenshots of a conductor card, such as the "Slack Code" conductor dropping from 28 chips, and "Reusable Issue Tracker Components" newly showing feat(app-sdk): issue list components (useIssues, IssueList, IssueCard) #34732 and feat(source): POST /api/source/issues lists and searches GitHub and GitLab issues #34742.
  • The "+N" expand. It now serves the owned list. To close: a screenshot of an expanded card.
  • Chip and tab qualifier text. KiroCrew and kirocrew now read as one project in the sidebar chips and the PullRequestPanel/IssuePanel tabs, which changes whether a repository label shows at all. To close: a screenshot of a mixed-case strip after the change.

Notes

  • A PR the chat only mentioned now disappears from the card with no on-surface explanation, and pinning one back is out of scope. Once the stills exist, check that a card's chips are still readable as "this chat's work" when nothing the user remembers seeing remains.

[UX-REVIEWED] 948d472

Description read

[DESCRIPTION-READ] 12953ca462f5edad59d67e9f9dc9ee9601b1ee04bf9cfb7ce94963ac9ae9f32c

That sha256 names everything this verdict read: the PR description, plus the 3 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 948d4729a14048ff32c95fd147782cd41e2e4b13 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: the ownership signals are additive, the chip payload and routes keep their shapes, and a revert brings back the old mention-scan chips.

Notes

  • ownership.py adds a 1,410-line parser that re-reads shell command text on every read and comes with a long list of known limits. tool_results.py already sees each gh-write result as it lands, so it could record the owned PR identity on the row once, in meta. The sidebar would then read that stored identity instead of parsing the command again.
  • agent_sdk now exposes the batch guard's private _gh_is_write, _GH_VALUED_OPTIONS and _positionals as public helpers. That lets a dashboard chip heuristic depend on internals of a security classifier. A change to the guard will now also change which chips show, so pin that coupling with a parity test on the guard side.
  • The description is cut off inside ## Backwards compatibility, so I couldn't check the Reader: list past the cut.

[DESIGN-REVIEWED] 948d472

Description read

[DESCRIPTION-READ] 1046611c83b7634db38c4c97b787e6bd6f9a3aa3a417412775c7d85a1e88e1bc

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — 🔴 BLOCK

Premise-level review of 948d4729a14048ff32c95fd147782cd41e2e4b13 — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: BLOCK

ownership.py writes its own wrapper-stepping (_command_start, _WRAPPERS, _is_assignment, _RESERVED, _is_wrapper_arg). security.argv_rules._program_args already does this job, and the batch guard uses it, so the card's reading of a command can drift from the guard's.

What this change ships

Inventory (10 items) — 8 justified

Intent: a card's GitHub chips show the PRs and issues the chat itself opened, acted on or reported, not every link it mentions. FIX.

  1. A PR or issue link only mentioned in a message no longer makes a chip — justified
  2. A successful gh pr/issue create row makes a chip — justified
  3. A successful gh write on a PR (edit, merge, api write) makes a chip — justified
  4. The chat's work-ledger report (pr, artifacts) makes a chip — justified
  5. A conductor's card shows its bound workers' created or reported PRs — justified
  6. Chips are ordered by the newest owning event, so a compaction summary cannot reorder them — justified
  7. Each tool row now stores its terminal status, and a restored tail keeps a gh-write row's output — justified
  8. A host-less owner/repo reads as github.com only after host settings load — justified
  9. Reading which program a gh call runs, past wrappers and NAME= — duplicate of security/argv_rules.py:_program_args
  10. KiroCrew and kirocrew count as one tab qualifier — rides along

Blockers

A second wrapper-and-assignment stepper: duplicate
Trigger: the description says the card reads gh calls "so the card and the guard agree". But ownership.py:61 adds _WRAPPERS = frozenset({"env", "command", "exec", "time", "nohup", "timeout", "nice", "stdbuf"}) alongside the guard's argv_rules._WRAPPERS (line 39). I grepped for wrapper tables in src/kiro_crew and found 3 in core: argv_rules, liveness_interactive.py:149, and ownership.
Outcome: the two already differ. The new one lacks sudo, setsid, ionice and builtin, and it skips any duration-shaped word after any word. So sudo gh pr create or setsid gh pr merge is a write the guard sees but the card misses.
Subtraction: delete _command_start, _WRAPPERS, _is_assignment, _RESERVED and _is_wrapper_arg. Find the program with argv_rules._program_args, and take the prefix for GH_HOST as the words before the arguments it returns.
Clears when: ownership.py steps over wrappers only through security.argv_rules._program_args.

[FIRST-PRINCIPLES-REVIEWED] 948d472

Description read

[DESCRIPTION-READ] 998140887fba5e12c4792de21e4e1b8fe641db668c69a54bd52232ff8af50683

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 948d4729a14048ff32c95fd147782cd41e2e4b13 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 948d472

Verdict parsed from the review's SHA-scoped output markers for commit 948d4729a14048ff32c95fd147782cd41e2e4b13.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 948d4729a14048ff32c95fd147782cd41e2e4b13 false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings (all downgraded on adjudication)

GPT 6.1 flagged blocking issues on 948d4729a14048ff32c95fd147782cd41e2e4b13; Opus 5.5 adjudication downgraded every one of them. Downgraded: the remedy each one needs is disproportionate to the harm it prevents. Adjudication downgraded all 1 blocking finding(s): the remedy each one requires is disproportionate to the harm it prevents.

This comment is updated in place on each push.

Review details

BLOCKING -- src/kiro_crew/dashboard/source_providers/ownership.py:1031 -- Successful creates claim PRs printed by unrelated reads

create_proven = all(call.decisive for call in create_calls) or not other_gh_prints
Trigger: gh pr create --fill && gh pr list --json url --jq '.[].url' succeeds, creating #10 and listing existing #9. SlotProjection.source_links → owned_refs_in_tool_row → _owned_refs labels both URLs "created" because the target-less list supplies no exclusion.
Outcome: Unrelated #9 appears on the chat’s card, consumes a visible chip position, and rolls up to its conductor as a PR the worker created.
Anchor: defect
Fix: Require unambiguous output provenance for create attribution even when the create succeeded; leave ambiguous URLs unowned while retaining independently proven explicit write targets.
Sites: src/kiro_crew/dashboard/source_providers/ownership.py:1060 also credits issue-create output; src/kiro_crew/dashboard/state_parts/source_links.py:501 rolls these links up, src/kiro_crew/dashboard/state_parts/source_links.py:544 exposes them through expansion, and src/kiro_crew/dashboard/state_parts/source_links.py:617 includes them in status refresh.
[BLOCK-MERGE-DOWNGRADED] 948d472
[GPT-REVIEWED] 948d472

Adjudication (Opus 5.5) — is blocking on each finding proportionate?

F1 — LOW harm. I confirmed every condition the finding needs:

  • Conditions: the create in a && chain is decisive, so create_proven holds (ownership.py:1031). The row must have succeeded (ownership.py:984). gh pr list names no target, so elsewhere and elsewhere_numbers exclude nothing (ownership.py:1046-1047). Every https:// line in the last 64 lines of output is then marked "created" (ownership.py:1048-1058). The code's own comment at ownership.py:1042-1044 already accepts this case as a known shortcut.

  • What happens when it fires: the dashboard shows one wrong chip, and the same link rolls up to the conductor (source_links.py:501). It is a visible display error rebuilt from the transcript on every read. Nothing is stored and no user data is lost. The trigger, a successful gh pr create and a URL-printing gh read in one shell command, is rare.

  • Cost of the real fix: splitting the output per gh call (the remedy that comment names), or new attribution logic for ambiguous URLs inside the create path. Either adds lasting complexity that costs more than an occasional wrong chip.

    [ADJUDICATION] 948d472 total=1 uphold=0 downgrade=1
    DOWNGRADE F1 src/kiro_crew/dashboard/source_providers/ownership.py:1031 reason=disproportionate-remedy
    [GPT-ADJUDICATED] 948d472

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 948d4729a14048ff32c95fd147782cd41e2e4b13 false-positive: <reason, 20 to 500 characters>

@KiroCrewCaptain

KiroCrewCaptain commented Nov 11, 2026 •

Copy link
Copy Markdown
Collaborator

Code change approver recommendation: A person should decide

Why: the decider raised 1 possible problem and estimates a 60% chance that this change introduces a defect; a person should check it.

The change replaces mention-based GitHub chips with ownership signals: gh write rows that exited 0, the work ledger, and roll-up from workers. The dashboard-turn path stores tool output and status and is well tested. But chats bound to a channel (Slack thread) persist tool rows without output or status, and their reply mentions no longer count, so their self-created PRs lose the chip they show on main. This regression on a sibling path isn't acknowledged as a non-goal and has no test. Smaller concerns (the copied gh api flag table, gh.exe) are non-blocking.

Things to check (possible problems the decider raised; at this confidence many turn out to be fine):

  • For a chat bound to a channel (Slack thread), a gh pr create the agent runs and announces in its reply made a chip on main through the reply's mention. On this head the mention is ignored and the tool row can never prove success, so the card shows no chip for PRs the chat created unless they were (src/kiro_crew/dashboard/channel_live.py, medium)

An automated recommendation, not a review: a maintainer still decides. It approves on its own only when the estimated chance of a defect is at most 12% and the change has at most 400 changed lines. Decider v3c (217fe607) · commit 9578cfbf7b · 3948 changed lines. If this recommendation is wrong, reply here: replies are how the approver is evaluated.

@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch 2 times, most recently from 19d100e to 3de23ff Compare November 11, 2026 17:13
@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Nov 11, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch 2 times, most recently from 6501205 to bea97a0 Compare November 12, 2026 00:24
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 12, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch from bea97a0 to f7c8ebb Compare November 12, 2026 02:29
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 12, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch from f7c8ebb to f0e520d Compare November 12, 2026 04:49
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Nov 12, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch 2 times, most recently from 6681d5c to bff19fc Compare November 12, 2026 07:00
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Nov 12, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch from 59d1fa3 to 0753823 Compare November 12, 2026 16:47
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 12, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch from 0753823 to 971bbc7 Compare November 12, 2026 19:16
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 12, 2026
@kyleseaman
kyleseaman force-pushed the fix/github-link-detection branch from 971bbc7 to 9578cfb Compare November 13, 2026 19:50
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Nov 13, 2026
Kyle Seaman added 2 commits November 13, 2026 21:10
…y PR it mentions

A session card's GitHub chips came from any pull-request or issue URL in any
row's text: status notes, pasted context and compaction summaries all made
chips, while the URL gh pr create printed (it sits in the tool row's output),
the PR a worker reported, and the PRs a conductor's workers opened never did.

A GitHub link now counts for a card when the chat created it (a successful
gh pr create / gh issue create row), acted on it (a successful gh pr
edit/comment/close/merge/ready/review or a writing gh api call on the pull),
reported it (its work-ledger item's pr or an artifacts PR URL), or a worker
bound to one of its items created or reported it (one level). Chips are ordered
by the newest owning event, so a compaction summary can no longer reshuffle
them. Other providers keep their mention rule, and the chat view's own Changes
and Resources panels are unchanged. The sidebar's owner/repo qualifier now
folds GitHub case like the backend identity already does.
…e shared shell reader, satisfy the repo guards
@kyleseaman

Copy link
Copy Markdown
Contributor Author

Closing without merging. The remaining work is tracked in #38578, the main issue, which links the branches, including an unpushed newer head on backup/fix-github-link-detection-be48090586. Follow-ups: #38579 (known shell-reading gaps), #38580 (pin a PR to a chat card), #38581 (status for chips in the expanded list). The branch is kept.

@kyleseaman kyleseaman closed this Nov 17, 2026
@github-actions github-actions Bot removed the readiness: action required A blocking check or review needs attention label Nov 17, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

merge conflict Branch has merge conflicts with its base — author must resolve before merge

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants