How real projects brief GitHub Copilot, from .github/copilot-instructions.md.
strausmannCopilot instructionsstrausmann/mcp-dockhand/.github/copilot-instructions.md
github/workflows/api-schema-sync.yml`). **If a tool's implementation differs from upstream, flag it as HIGH priority.** ## Code Review Guidelines ### Security - All path parameters MUST use `encodePath()` from `src/utils/encode-path.ts` - No hardcoded credentials
0xmariowuCopilot instructions0xmariowu/Autosearch/.github/copilot-instructions.md
with a TypeScript CLI wrapper and npm publishing pipeline. Any Copilot agent (Coding Agent, Code Review, ci-doctor) working on this repo must follow the rules below. ## Hard rules — never
MicrosoftCopilot instructionsmicrosoft/devsquad-copilot/.github/copilot-instructions.md
recommendation to do nothing. - Generating unnecessary code or changes is worse than generating nothing. - When asked to "analyze", "review", or "consider", the result can legitimately be: "I analyzed
MicrosoftCopilot instructionsmicrosoft/devsquad-copilot/.github/plugins/devsquad/hooks/templates/.github/copilot-instructions.md
recommendation to do nothing. - Generating unnecessary code or changes is worse than generating nothing. - When asked to "analyze", "review", or "consider", the result can legitimately be: "I analyzed
MicrosoftCopilot instructionsmicrosoft/duroxide-pg/.github/copilot-instructions.md
changed function must be shown with `+`/`-` diff markers on changed lines inside a ` ```diff ` code block. The diff file should contain: 1. **Table Changes** — New tables (full
ssdeanxCopilot instructionsssdeanx/AgentStack/.github/copilot-instructions.md
understand and contribute to the codebase. - 🧑💻 Before making any changes to the code, take the time to review the existing codebase and understand the context and dependencies of the code
Robotti-ioCopilot instructionsRobotti-io/copilot-security-instructions/copilot-instructions.md
with secrets or file I/O — it’s probably unsafe. - Treat AI-generated code as a draft; always review and test before integration. - Maintain a human-in-the-loop approach
Robotti-ioCopilot instructionsRobotti-io/copilot-security-instructions/.github/copilot-instructions.md
instructions: `instructions/*.instructions.md` - Frontmatter MUST include: - `applyTo: "<glob pattern(s)>"` - Frontmatter MAY include: - `excludeAgent: "code-review"` or `"coding-agent"` if only one should read it - Body MUST: - Describe exactly what
jmrplensCopilot instructionsjmrplens/gitlab-mcp-server/.github/copilot-instructions.md
agents: `go-mcp-expert` (primary coding), `test-expert` (testing, coverage, false-pass detection), `plan-expert` (strategic planning), `debug` (debugging), `se-reviewer` (OWASP + architecture), `documentation-writer` (project docs with Context7
MicrosoftCopilot instructionsmicrosoft/Agent365-python/.github/copilot-instructions.md
GitHub Copilot Instructions for Agent365-python ## Code Review Rules ### Rule 1: Check for "Kairo" Keyword - **Description**: Scan code for any occurrence of the keyword "Kairo" - **Action**: If "Kairo" is found
anthony-chaudharyCopilot instructionsanthony-chaudhary/fak/.github/copilot-instructions.md
clean** — use guarded headless agents or equivalent isolated workers for investigation, implementation, tests, and review. Keep only decisions and compact witnessed evidence in the primary context; independently verify worker effects
intersystems-communityCopilot instructionsintersystems-community/iris-agentic-dev/.github/copilot-instructions.md
mocked Atelier client, no stubbed responses. A mocked IRIS test passes while the real code path is broken, which is worse than no test. Integration and e2e runs need `--test
MicrosoftCopilot instructionsmicrosoft/durabletask-python/.github/copilot-instructions.md
implemented internally). - Changelogs are not covered by the CI Markdown lint step. Review changes to them manually. - Use the current unindented changelog style: category labels such as `ADDED`, `CHANGED
personamanagmentlayerCopilot instructionspersonamanagmentlayer/pcl/.github/copilot-instructions.md
feature] architecture following src/ patterns # Implementation /persona DEV - Implement [feature] with tests following copilot-instructions.md # Code review /persona CRITIC - Review changes against quality standards # Security audit /persona SEC - Security review
girishrCopilot instructionsgirishr/SpecPilot/.github/copilot-instructions.md
next step seems obvious, ask first — do not assume. 8. **SPEC-FIRST review gate**: Before touching any code or non-spec files, read all relevant `.specs/` files, update all affected
MicrosoftCopilot instructionsmicrosoft-foundry/forgebook/.github/copilot-instructions.md
cookbook publishing platform. Jupyter notebooks are the source of truth; metadata lives in `registry.yaml`. ## Code Review Focus Areas When reviewing code in this repository: - **Notebooks are sacred** — Never add site
padmarajnidagundiCopilot instructionspadmarajnidagundi/Playwright-AI-Agent-POM-MCP-Server/.github/copilot-instructions.md
repo - `.github/skills/playwright-test-debugging/SKILL.md` — auto-loaded when debugging failing tests - `.github/skills/code-review/SKILL.md` — auto-loaded when reviewing, auditing, or inspecting test code; enforces POM compliance, Playwright best practices, security rules, and coverage completeness Hints
sushichan044Copilot instructionssushichan044/cc-hooks-ts/.github/copilot-instructions.md
Copilot Code Review Instructions `cc-hooks-ts` is a TypeScript library that defines Claude Code hook types with runtime validation via `valibot`. It tracks upstream `@anthropic-ai/claude-agent-sdk` and Claude
MicSmCopilot instructionsMicSm/boffin/.github/copilot-instructions.md
refactor breaks the task boundary and gets rejected. - Open-ended refactor, cleanup, or review of existing code: run the two-pass audit below, scoped to what the task covers
ElasticCopilot instructionselastic/opentelemetry/.github/copilot-instructions.md
pages to preserve SEO value - Review user feedback to identify unclear sections that need improvement - Update screenshots and UI references when interfaces change - Refresh code examples to reflect current product
Repository-wide instructions for GitHub Copilot, kept in .github/copilot-instructions.md.
How to build and test, the project's structure, and the conventions Copilot should follow in suggestions and reviews.
No. Copilot's coding agent also reads AGENTS.md, so many projects keep the shared rules there.
Open a file to see its discussion. Reports from people and their agents are coming.
