fix: power-steering incorrectly blocks Q&A/PM sessions with dev checks (#2914)#2916
Closed
fix: power-steering incorrectly blocks Q&A/PM sessions with dev checks (#2914)#2916
Conversation
#2914) - Remove INVESTIGATION from workflow_invocation, next_steps, and documentation_updates applicable_session_types — these checks are designed for development sessions and produce false positives on investigation/PM sessions - Add new skill_invocation check: if a session starts with a skill command (<command-name> tag), verify the Skill tool was actually called for that skill. This catches cases where Claude bypasses a requested skill without blocking sessions that correctly use skills. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
|
🤖 Auto-fixed version bump The version in If you need a minor or major version bump instead, please update |
Owner
Author
|
Closing: merge conflicts with the power_steering_checker package refactor (#2910). The considerations.yaml and checks_workflow.py changes will be reimplemented in a fresh PR against the current package structure. The skill_invocation check is valuable and will be preserved. |
3 tasks
rysweet
pushed a commit
that referenced
this pull request
Mar 7, 2026
Detects when a user requests a skill via slash command (<command-name> tag) but the agent bypasses it and responds directly without invoking the Skill tool. This was independently valuable work from closed PR #2916 that got dropped during the stale PR cleanup. - checks_workflow.py: Add _check_skill_invocation method - considerations.yaml: Add skill_invocation blocker for DEV/INVESTIGATION/MAINTENANCE - 12 outside-in tests (Claude + Copilot + edge cases) Related to #2914 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
3 tasks
rysweet
added a commit
that referenced
this pull request
Mar 7, 2026
…2926) * feat: add skill_invocation power-steering check (extracted from #2916) Detects when a user requests a skill via slash command (<command-name> tag) but the agent bypasses it and responds directly without invoking the Skill tool. This was independently valuable work from closed PR #2916 that got dropped during the stale PR cleanup. - checks_workflow.py: Add _check_skill_invocation method - considerations.yaml: Add skill_invocation blocker for DEV/INVESTIGATION/MAINTENANCE - 12 outside-in tests (Claude + Copilot + edge cases) Related to #2914 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * [skip ci] chore: Auto-bump patch version --------- Co-authored-by: Ubuntu <azureuser@deva.ftnmxvem3frujn3lepas045p5c.xx.internal.cloudapp.net> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
workflow_invocation,next_steps, anddocumentation_updatesapplicable_session_types — these are dev-specific checks that produce false positives on investigation/PM sessionsskill_invocationcheck: if a<command-name>tag exists in the transcript (user ran a slash command), verify the Skill tool was called for that skillProblem
During a
/pm-architectsession (pure PM prioritization), power-steering blocked the session 4-5 times with irrelevant checks likeworkflow_invocation(demanded dev-orchestrator),next_steps(flagged the prioritized backlog as "remaining work"), anddocumentation_updates.Root cause
Three INVESTIGATION blocker checks were designed for development workflows and don't apply to PM/research sessions:
workflow_invocation— demands Skill/Read for workflow file, but PM skills don't need dev-orchestratornext_steps— flags project backlogs as incomplete work when the backlog IS the deliverabledocumentation_updates— demands doc updates for sessions that don't change codeTest plan
/pm-architectand verify power-steering doesn't block with dev checks/devand verifyskill_invocationcheck passes when dev-orchestrator is called🤖 Generated with Claude Code