Skip to content

Stabilize Step Name Alignment by permitting direct manifest reads and tightening agent guidance#34873

Merged
pelikhan merged 2 commits into
mainfrom
copilot/aw-step-name-alignment-fix
May 26, 2026
Merged

Stabilize Step Name Alignment by permitting direct manifest reads and tightening agent guidance#34873
pelikhan merged 2 commits into
mainfrom
copilot/aw-step-name-alignment-fix

Conversation

Copy link
Copy Markdown
Contributor

Copilot AI commented May 26, 2026

The Step Name Alignment workflow was failing in agent with error_max_turns after repeated denied Bash attempts against /tmp/gh-aw/agent and /tmp/gh-aw/cache-memory. This change aligns tool permissions and prompt instructions so the agent can consume the prebuilt manifest without exploratory /tmp probing loops.

  • Workflow permission alignment

    • Added explicit Bash allowlist entries in .github/workflows/step-name-alignment.md for direct manifest access:
      • cat /tmp/gh-aw/agent/step-alignment-input.json
      • jq* /tmp/gh-aw/agent/step-alignment-input.json
  • Prompt hardening to avoid denial loops

    • Updated instructions to use the cache-memory tool directly instead of shell exploration of /tmp/gh-aw/cache-memory.
    • Added explicit manifest-read command and guidance to avoid ls/find/cp discovery scans in /tmp.
  • Compiled workflow sync

    • Recompiled .github/workflows/step-name-alignment.lock.yml so runtime allowed-tools reflect the new manifest-read permissions.
tools:
  bash:
    - yq*
    - find .github/workflows -name "*.lock.yml" -type f
    - cat docs/src/content/docs/reference/glossary.md
    - cat /tmp/gh-aw/agent/step-alignment-input.json
    - jq* /tmp/gh-aw/agent/step-alignment-input.json

Copilot AI linked an issue May 26, 2026 that may be closed by this pull request
Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
Copilot AI changed the title [WIP] Fix step name alignment issues Stabilize Step Name Alignment by permitting direct manifest reads and tightening agent guidance May 26, 2026
Copilot AI requested a review from pelikhan May 26, 2026 05:35
@pelikhan pelikhan marked this pull request as ready for review May 26, 2026 05:38
Copilot AI review requested due to automatic review settings May 26, 2026 05:38
@pelikhan pelikhan merged commit cc2e15f into main May 26, 2026
@pelikhan pelikhan deleted the copilot/aw-step-name-alignment-fix branch May 26, 2026 05:38
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

This PR aims to stop the Step Name Alignment workflow from hitting error_max_turns by (1) granting the agent explicit permission to read the prebuilt step-alignment manifest and (2) tightening prompt guidance to avoid denied /tmp probing loops.

Changes:

  • Added explicit bash allowlist entries to read /tmp/gh-aw/agent/step-alignment-input.json (via cat and jq).
  • Updated agent instructions to prefer the cache-memory tool and avoid exploratory /tmp scans.
  • Recompiled the workflow lock file so the runtime allowed-tools list reflects the new permissions.
Show a summary per file
File Description
.github/workflows/step-name-alignment.md Expands agent bash allowlist and updates instructions around cache-memory + manifest consumption.
.github/workflows/step-name-alignment.lock.yml Regenerated compiled workflow to include the new allowed-tools entries.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

  • Files reviewed: 2/2 changed files
  • Comments generated: 1

Read it directly with:

```bash
cat /tmp/gh-aw/agent/step-alignment-input.json | jq .
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[aw] Step Name Alignment failed

3 participants