agentic-coder-skill
Copilot instructions2 starsChanged 2 months agoApache-2.0View on GitHub
Copilot has a 2-iteration limit per PR: 1. Initial implementation 2. One round of review feedback If feedback requires more than one iteration, close the PR and create a new issue with updated requirements. When reviewing Copilot PRs, verify: Copilot runs in a firewalled GitHub Actions environment. If your project requires external URLs (APIs, registries, artifact stores), document them here so Copilot can request network access: Copilot code review is enforceable via the copilot_code_review ruleset rule (added to the branch…
Where files are headingNew Copilot instructions files have more rules about branches and pull requests, and fewer about code style.
Of the Copilot instructions files we read, the 2,349 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
The file
We split this copilot-instructions.md into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Use the Assignees field on the issue page.
- Distinctive: line 22
If feedback requires more than one iteration, close the PR and create a new issue with updated requirements.
- Distinctive: line 26
When reviewing Copilot PRs, verify:
- Very specific: line 31
SHA pinning: If the PR modifies .github/workflows/*.yml, all action uses: must be SHA-pinned with a comment showing the tag (e.g., uses: …
- Distinctive: line 32
Commits: All commits must be GPG-signed and include DCO sign-off (git commit -S --signoff).
≈ “Common” is an estimate: at least 1 in 5 of the 3,779 Copilot instructions files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Copilot Instructions
Assigning Issues and Triggering Rework
Assign via REST API
gh issue edit <NUMBER> --add-assignee copilot-swe-agentAssign via GitHub UI
Use the Assignees field on the issue page.
Trigger Rework
Mention @copilot in a PR comment (GitHub UI only, not via gh). Copilot will re-run and amend the PR.
PR Iteration Limit
Copilot has a 2-iteration limit per PR:
- Initial implementation
- One round of review feedback
If feedback requires more than one iteration, close the PR and create a new issue with updated requirements.
Code Review Checklist
When reviewing Copilot PRs, verify:
- Scope creep: Does the PR implement only what the issue specifies?
- API correctness: Are all library/framework calls valid? Verify against installed package versions or docs.
- Error handling: Are errors caught and handled appropriately for the context?
- SHA pinning: If the PR modifies
.github/workflows/*.yml, all actionuses:must be SHA-pinned with a comment showing the tag (e.g.,uses: owner/action@<SHA> # vX.Y.Z). - Commits: All commits must be GPG-signed and include DCO sign-off (
git commit -S --signoff).
Firewall Constraints
Copilot runs in a firewalled GitHub Actions environment. If your project requires external URLs (APIs, registries, artifact stores), document them here so Copilot can request network access:
<!-- Add any external URLs needed by the project -->Enabling Copilot code review
Copilot code review is enforceable via the copilot_code_review ruleset rule (added to the
branch ruleset). Once enabled, Copilot automatically reviews every PR on push.
To request a re-review: re-request review from Copilot in the PR sidebar.
Design References
For detailed standards, see:
- CONTRIBUTING.md
- docs/REPO-STANDARDS.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

