Skip to content

fix(#1965): clarify the /fs-code hint after triage auto-promotion - #2040

Open
Benkapner wants to merge 2 commits into
mainfrom
fix/1965-triage-auto-code-hint
Open

Benkapner wants to merge 2 commits into
mainfrom
fix/1965-triage-auto-code-hint

Conversation

@Benkapner

Copy link
Copy Markdown
Member

Summary

When triage adds ready-to-code, the label can start a code run automatically. The triage comment still told the maintainer to post /fs-code. In the case reported in #1965, following that hint cancelled the active run and started a second one.

Related Issue

Closes #1965.

Changes

  • When triage adds ready-to-code, explain that repositories with automatic dispatch start the code agent from that label.
  • Describe /fs-code <your instruction> as a way to start a new run with different instructions.
  • Keep the existing manual start hint for sufficient issues that triage does not promote.

Testing

  • Two regression checks failed against the original script. Both pass with the fix.
  • bash scripts/post-triage-test.sh --bundled passed all 305 checks.
  • make test
  • make script-build and make check-bundle
  • make lint, pre-commit run --all-files, and git diff --check

The ten-issue production check in #1965 is a follow-up after rollout.

Checklist

  • PR title follows Conventional Commits.
  • Commit is signed off (DCO).
  • I wrote this contribution myself and can explain all changes in it.

When triage adds ready-to-code, following the manual start hint can
cancel the code run that the label already triggered. Explain the
automatic path and reserve the command for a deliberate new run with
different instructions. Keep the manual hint for other sufficient issues.

Signed-off-by: Ben Kapner <benkapner@gmail.com>
@Benkapner Benkapner self-assigned this Nov 15, 2026
@Benkapner
Benkapner requested a review from a team as a code owner November 15, 2026 16:19
@qodo-code-review

Copy link
Copy Markdown

PR Summary by Qodo

Clarify code-run hints after triage auto-promotion

🐞 Bug fix 🧪 Tests 🕐 10-20 Minutes

Grey Divider

AI Description

• Explain that ready-to-code can start a code run automatically, avoiding a redundant /fs-code
 command.
• Present /fs-code  as a deliberate new run; retain manual-start guidance otherwise.
• Add regression checks for both footer paths.
Diagram

graph TD
  A["Sufficient triage"] --> B{"Ready-to-code deferred?"} -->|Yes| C["Auto-dispatch hint"] --> E["Sticky comment"]
  B -->|No| D["Manual-start hint"] --> E
Loading
High-Level Assessment

Branching on the deferred ready-to-code label is appropriate because it reflects the triage decision that makes the manual-start hint misleading. Branching on configuration alone would not account for issues that remain unpromoted despite automatic dispatch being enabled.

Files changed (3) +25 / -8

Bug fix (2) +16 / -4
post-triage.shSelect next-steps guidance by deferred label +8/-2

Select next-steps guidance by deferred label

• The bundled script now explains automatic dispatch when triage defers 'ready-to-code' and reserves '/fs-code <your instruction>' for a new run. Other sufficient issues keep the existing manual-start guidance.

scripts/post-triage.sh

post-triage.src.shCorrect the triage comment footer in source +8/-2

Correct the triage comment footer in source

• Selects auto-dispatch or manual-start next steps according to whether triage deferred 'ready-to-code', preventing an auto-promoted issue from recommending a redundant command.

scripts/post-triage.src.sh

Tests (1) +9 / -4
post-triage-test.shCover auto-promoted and manual-start footer guidance +9/-4

Cover auto-promoted and manual-start footer guidance

• Replaces the broad '/fs-code' footer check with checks that auto-promoted issues explain the label trigger and omit the manual-start hint. Verifies that disabling auto-code retains the manual-start command.

scripts/post-triage-test.sh

@qodo-code-review

qodo-code-review Bot commented Nov 15, 2026 •

Copy link
Copy Markdown

Code Review by Qodo

🐞 Bugs (0) 📘 Rule violations (0) 📎 Requirement gaps (1) 🔗 Cross-repo conflicts (0) 📜 Skill insights (0)

Grey Divider


Remediation recommended

1. Footer fails to say code run started 📎 Requirement gap ≡ Correctness
Description
The ready-to-code footer in scripts/post-triage.src.sh begins with the conditional phrase "With
automatic dispatch enabled ... starts the code agent", even though that branch only runs after
AUTO_CODE_ALLOWED is true and DEFERRED_LABEL is set, so dispatch is already known. Because the
wording is conditional and future-tense, maintainers cannot tell whether a run has started and may
invoke /fs-code, cancelling it; the same text is copied into the generated
scripts/post-triage.sh.
Code

scripts/post-triage.src.sh[R894-896]

+  if [[ "${DEFERRED_LABEL}" == "ready-to-code" ]]; then
+    NEXT_STEPS='- With automatic dispatch enabled, the `ready-to-code` label starts the code agent. No command is needed.
+- To start a new run with different instructions, use `/fs-code <your instruction>`.'
Relevance

●●● Strong

Wording fix matching the PR's stated intent to clarify the footer; team accepts clarity rewording of
trigger docs (PR 389). No rejection precedent.

PR-#389

ⓘ Recommendations generated based on similar findings in past PRs

Evidence
Checklist item 4134520 requires the footer, when ready-to-code is applied, to state that the code
agent has already been dispatched and to offer /fs-code <your instruction> only for added guidance
or a restart. The current text is conditional ("with automatic dispatch enabled ... starts"),
leaving an unresolved 'if' that the script has already resolved, so it reads as a guess rather than
fact. It also does not tell a maintainer what to do if the run does not appear, and the requirement
is therefore not met.

Acknowledge automatic code dispatch in the sufficient triage footer
scripts/post-triage.src.sh[894-896]
scripts/post-triage.src.sh[894-899]
scripts/post-triage.sh[2038-2049]

Agent prompt
The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
The sufficient-triage footer in `scripts/post-triage.src.sh` (and the generated `scripts/post-triage.sh`) uses conditional wording, "With automatic dispatch enabled ... starts the code agent", when the `ready-to-code` label is applied. That branch only runs after `AUTO_CODE_ALLOWED` is true and `DEFERRED_LABEL` is set, so dispatch is already known. The footer should state as fact that the code agent has already been dispatched, matching checklist item 4134520, so maintainers do not cancel a run by invoking `/fs-code`.

## Fix Focus Areas
- scripts/post-triage.src.sh[894-899]
- scripts/post-triage.sh[2360-2366]
- scripts/post-triage-test.sh[1054-1058]

## Recommended Fix
In the `ready-to-code` branch of `scripts/post-triage.src.sh`, replace the conditional opening with a direct statement such as `🚀 Code agent automatically dispatched to create a PR`. Keep the second bullet offering `/fs-code <your instruction>` only as a way to add guidance or start a new run. Optionally add a line on what to do if the run does not appear. Regenerate `scripts/post-triage.sh` from the source with `make script-build` and confirm with `make check-bundle`. Update the regression check in `scripts/post-triage-test.sh` so it asserts the dispatched wording rather than only the `ready-to-code` label phrase.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

Dismiss ↗ | View ↗


2. Footer still invites the command that cancels active runs ✓ Resolved

Resolving commit: 8e8bc4f

Resolution reason
The ready-to-code NEXT_STEPS text in post-triage.src.sh (and the generated post-triage.sh) now says
to use /fs-code <your instruction> only to start over with different instructions and explicitly
warns that this can cancel the current code run. A test in post-triage-test.sh was added to assert
the warning text. This addresses the finding's concern with different wording than the suggested fix
(which asked for a consistent clause mirroring the other branch), so it is not a direct
implementation.
Description
In the ready-to-code branch, NEXT_STEPS still advertises /fs-code <your instruction> as the
way to start a new run, with no warning that posting it while the automatic run is active cancels
that run. The ticket's incident shows the shared concurrency group with cancel-in-progress: true
makes this exact action produce a Cancelled card and a redundant run, so the footer keeps the trap
the PR is meant to close for maintainers who want different instructions.

Grey Divider

Context sources
✅ Compliance rules (platform): 55 rules
✅ Skills: 4 invoked
  code-review
  code-implementation
  pr-review
  docs-review
✅ Cross-repo context — repo relationships
Review mode: Auto: 🚀 Fast: Localized footer messaging change with contained, well-tested behavior.

Grey Divider

Tip of the day
💡 Did you know, you can reply 'qodo' on any finding to push back, ask questions, or dig deeper

More tips ↗ | Customize Qodo ↗ | Qodo docs ↗

Grey Divider

Qodo Logo

Comment thread scripts/post-triage.src.sh Outdated
Comment on lines +894 to +896
if [[ "${DEFERRED_LABEL}" == "ready-to-code" ]]; then
NEXT_STEPS='- With automatic dispatch enabled, the `ready-to-code` label starts the code agent. No command is needed.
- To start a new run with different instructions, use `/fs-code <your instruction>`.'

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Remediation recommended

1. Footer fails to say code run started 📎 Requirement gap ≡ Correctness

The ready-to-code footer in scripts/post-triage.src.sh begins with the conditional phrase "With
automatic dispatch enabled ... starts the code agent", even though that branch only runs after
AUTO_CODE_ALLOWED is true and DEFERRED_LABEL is set, so dispatch is already known. Because the
wording is conditional and future-tense, maintainers cannot tell whether a run has started and may
invoke /fs-code, cancelling it; the same text is copied into the generated
scripts/post-triage.sh.
Agent Prompt
## Issue description
The sufficient-triage footer in `scripts/post-triage.src.sh` (and the generated `scripts/post-triage.sh`) uses conditional wording, "With automatic dispatch enabled ... starts the code agent", when the `ready-to-code` label is applied. That branch only runs after `AUTO_CODE_ALLOWED` is true and `DEFERRED_LABEL` is set, so dispatch is already known. The footer should state as fact that the code agent has already been dispatched, matching checklist item 4134520, so maintainers do not cancel a run by invoking `/fs-code`.

## Fix Focus Areas
- scripts/post-triage.src.sh[894-899]
- scripts/post-triage.sh[2360-2366]
- scripts/post-triage-test.sh[1054-1058]

## Recommended Fix
In the `ready-to-code` branch of `scripts/post-triage.src.sh`, replace the conditional opening with a direct statement such as `🚀 Code agent automatically dispatched to create a PR`. Keep the second bullet offering `/fs-code <your instruction>` only as a way to add guidance or start a new run. Optionally add a line on what to do if the run does not appear. Regenerate `scripts/post-triage.sh` from the source with `make script-build` and confirm with `make check-bundle`. Update the regression check in `scripts/post-triage-test.sh` so it asserts the dispatched wording rather than only the `ready-to-code` label phrase.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

Dismiss ↗ | View ↗

Comment thread scripts/post-triage.src.sh Resolved
@fullsend-ai-review

fullsend-ai-review Bot commented Nov 15, 2026 •

Copy link
Copy Markdown

🤖 Finished Review · ✅ Success · Started 4:21 PM UTC · Completed 4:27 PM UTC

Commit: d506e3a · View workflow run →

Runtime: pi · Model: sonnet → claude-sonnet-5-5 · Effort: high · Cost: $1.25

@fullsend-ai-review fullsend-ai-review Bot added the risk/moderate PR risk: moderate label Nov 15, 2026
@fullsend-ai-review

fullsend-ai-review Bot commented Nov 15, 2026 •

Copy link
Copy Markdown

Risk Assessment: moderate (2/5)

Details

Small 36-line triage footer update in protected scripts path with elevated churn and regression history, but tight issue alignment and accompanying unit tests keep risk moderate.

Previous run

Risk Assessment: moderate (2/5)

Details

Small 33-line triage footer update in protected scripts path with elevated churn and regression history, but tight issue alignment and accompanying unit tests keep risk moderate.

@fullsend-ai-review

fullsend-ai-review Bot commented Nov 15, 2026 •

Copy link
Copy Markdown

Review

Findings

Medium

  • [protected-path] scripts/post-triage.sh, scripts/post-triage.src.sh, scripts/post-triage-test.sh — This PR modifies files under the protected scripts/ path (governance/infrastructure). The PR links Acknowledge automated code dispatch in post-triage next-steps footer when ready-to-code is applied #1965 and explains the rationale (the triage footer told maintainers to post /fs-code after ready-to-code had already started a code run), so context is sufficient. Human approval is always required for protected-path changes, regardless of context.
    Remediation: A human maintainer with authority over scripts/ should review and approve this change.
Previous run

<!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJhY3Rpb24iOiJjb21tZW50IiwiZmluZGluZ3MiOltdfQ== -->

Review

Findings

Medium

  • [protected-path] scripts/post-triage.sh, scripts/post-triage.src.sh, scripts/post-triage-test.sh — This PR modifies files under the protected scripts/ path. The PR links issue Acknowledge automated code dispatch in post-triage next-steps footer when ready-to-code is applied #1965 (Closes #1965) and explains the rationale (the triage footer told maintainers to run /fs-code after ready-to-code had already auto-dispatched the code agent). Human approval is always required for protected-path changes, regardless of context.
    Remediation: A human maintainer must review and approve these protected-path changes.

Labels: PR fixes a bug in triage footer messaging

@fullsend-ai-review fullsend-ai-review Bot added requires-manual-review Review requires human judgment bug Something isn't working labels Nov 15, 2026
Signed-off-by: Ben Kapner <benkapner@gmail.com>
@fullsend-ai-review

fullsend-ai-review Bot commented Nov 15, 2026 •

Copy link
Copy Markdown

🤖 Finished Review · ✅ Success · Started 5:22 PM UTC · Completed 5:29 PM UTC

Commit: 8e8bc4f · View workflow run →

Runtime: pi · Model: sonnet → claude-sonnet-5-5 · Effort: high · Cost: $1.53

This branch was successfully deployed

1 active deployment
dev — 8e8bc4fd Deployed Nov 15, 2026 by Benkapner via behaviour #2151
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

bug Something isn't working requires-manual-review Review requires human judgment risk/moderate PR risk: moderate

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Acknowledge automated code dispatch in post-triage next-steps footer when ready-to-code is applied

1 participant