Agent Persona Exploration - 2026-04-30 #29241
Closed
Replies: 1 comment
-
|
This discussion has been marked as outdated by Agent Persona Explorer. A newer discussion is available at Discussion #29387. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Persona Overview
Key Findings
pull_requestwithpaths:filters,workflow_run,schedule, and other appropriate triggers (5/5 average)add-pr-comment,create-issue,create-discussionas appropriatepull-requests: write,discussions: write) instead of keeping the agent read-only and relying solely on safe-outputsTop Patterns
pull_request: types: [opened, synchronize]for PR-based workflowsgithub: toolsets: [default](DevOps scenario correctly used specific toolsets[actions, issues, repos])View High Quality Responses (Top 2)
DevOps — Deployment Incident Reporter (5.0/5.0)
workflow_run: types: [completed]trigger — correct and preciseactions: read, contents: read) ✅[actions, issues, repos]instead of generic[default]max: 1+expires: 1hrate limiting oncreate-issuesafe-output — preventing issue floodingBackend Engineer — DB Migration Review (4.4/5.0)
paths:filter onpull_requesttrigger to scope to SQL files onlyView Areas for Improvement
Critical: Write permissions on agent job (3/4 scenarios affected)
The agent repeatedly added write permissions directly to the agent job:
The correct pattern per
.github/aw/create-agentic-workflow.md:The DevOps scenario was the only one that got this right — likely because its write operation (create-issue) was clearly separated from reading GitHub data.
Minor: Generic toolset
[default]vs. specific toolsetsMost scenarios defaulted to
toolsets: [default]which works but is broader than necessary. Scoped toolsets (e.g.,[issues, pull_requests]) would follow least-privilege principles better.Recommendations
Strengthen the permission model guidance in
.github/aw/create-agentic-workflow.md— add a prominently-placed "permission quick-reference table" showing which safe-output requires which read permission (e.g.,add-pr-comment→pull-requests: read, neverwrite). The current guidance exists but agents still generate write permissions 75% of the time.Add a "safe-outputs permission mapping" section to
.github/aw/github-agentic-workflows.md— explicitly document that all safe-outputs operate independently of the agent job's permission scope, and that adding write permissions to the agent job is always wrong when safe-outputs are in use.Encourage specific GitHub toolsets over
[default]— the DevOps scenario demonstrated better least-privilege by using[actions, issues, repos]. Consider adding guidance to scope toolsets to only what the workflow needs.References:
Beta Was this translation helpful? Give feedback.
All reactions