AGENTS.md examples from real projects

lobehubReads credentialsAGENTS.md

lobehub

lobehub/lobehub/AGENTS.md

skill for placement, naming, test location, and how a large file becomes a directory. ### Code Review Before reviewing a PR / diff / branch change, read the **deep-review** skill. Ordinary review

bytedanceAGENTS.md

deer-flow

bytedance/deer-flow/AGENTS.md

edit a directory, read every `AGENTS.md` from that directory up to this file. Some coding agents do not load the nested files on their own. | Area | Guide | | --- | --- | | Backend: harness, Gateway

heygen-comAGENTS.md

hyperframes

heygen-com/hyperframes/docs/AGENTS.md

Never infer product behavior from page titles or old docs. Verify it in current code. - Do not preserve a page merely because it already exists. Merge, rewrite, redirect, or remove

No other project uses it yet

lobehubReads credentialsAGENTS.md

lobe-chat

lobehub/lobe-chat/AGENTS.md

skill for placement, naming, test location, and how a large file becomes a directory. ### Code Review Before reviewing a PR / diff / branch change, read the **deep-review** skill. Ordinary review

jesseduffieldAGENTS.md

lazygit

jesseduffield/lazygit/AGENTS.md

follow-up commit. This gives reviewers (and `git bisect`) a clear before/after and proves the test actually exercises the broken code path. This applies only to defects that existed before

No other project uses it yet

career-ops-hqReads credentialsAGENTS.md

career-ops

career-ops-hq/career-ops/AGENTS.md

career-ops project (parent/sibling repos, other codebases on the machine), knowledge from other Claude Code projects on the same machine, and cross-session inferences not written into an in-scope

career-ops-hqAGENTS.md

career-ops / web

career-ops-hq/career-ops/web/AGENTS.md

finds the user's files, and it keeps your real pipeline out of it. ## Reviews and scope `web/` is maintained as first-party: functional and correctness fixes are very welcome

netdataReads credentialsAGENTS.md

netdata

netdata/netdata/AGENTS.md

Review findings are leads until verified against the relevant design or shipped code and its contracts. - A review request alone is read-only, as defined under "When

No other project uses it yet

santiferReads credentialsAGENTS.md

career-ops

santifer/career-ops/AGENTS.md

career-ops project (parent/sibling repos, other codebases on the machine), knowledge from other Claude Code projects on the same machine, and cross-session inferences not written into an in-scope

ruvnetReads credentialsAGENTS.md

claude-flow

ruvnet/claude-flow/AGENTS.md

type`) | Type | Purpose | |------|---------| | `coordinator` | Orchestrates other agents | | `coder` | Writes code | | `tester` | Writes tests | | `reviewer` | Reviews code | | `architect` | Designs systems | | `researcher` | Analyzes requirements | | `security-architect` | Security design | | `performance-engineer` | Optimization | ### Task

storytoldAGENTS.md

photocraft

storytold/photocraft/AGENTS.md

agent answers for what it opens and for its review comments. 2. **Check, then claim.** Before writing code, read the issue comments, search pull requests for `#N` and the keywords

MicrosoftReads credentialsAGENTS.md

ai-agents-for-beginners

microsoft/ai-agents-for-beginners/AGENTS.md

Notebooks should execute without errors - README files should be clear and accurate - Follow existing code patterns in the repository - Maintain consistency with other lessons ## Additional Notes ### Common Gotchas 1. **Python

No other project uses it yet

MicrosoftReads credentialsAGENTS.md

ai-agents-for-beginners / en

microsoft/ai-agents-for-beginners/translations/en/AGENTS.md

Notebooks should execute without errors - README files should be clear and accurate - Follow existing code patterns in the repository - Maintain consistency with other lessons ## Additional Notes ### Common Gotchas 1. **Python

No other project uses it yet

MicrosoftReads credentialsAGENTS.md

ai-agents-for-beginners / id

microsoft/ai-agents-for-beginners/translations/id/AGENTS.md

example for ` - `[Fix] Correct typo in lesson-XX README` - `[Update] Improve code sample in lesson-XX` - `[Docs] Update setup instructions` ### Pemeriksaan Wajib - Notebook harus dijalankan tanpa error - Berkas README harus

No other project uses it yet

rohitg00Deletes or force-pushesAGENTS.md

ai-engineering-from-scratch

rohitg00/ai-engineering-from-scratch/AGENTS.md

interview loop for a role, or a review of interview gaps, read and follow `skills/interview-prep/SKILL.md` before asking anything. Claude Code also discovers the copy under `.claude/skills/`. The skill asks

caddyserverAGENTS.md

caddy

caddyserver/caddy/AGENTS.md

security-first, modular, extensible server platform. ## Code Style ### Go Idioms Follow [Go Code Review Comments](https://go.dev/wiki/CodeReviewComments): - **Error flow**: Early return, indent error handling—not else blocks

No other project uses it yet

ElasticAGENTS.md

elasticsearch

elastic/elasticsearch/AGENTS.md

breaker reservations, pooled memory allocations, etc) must be released eventually. - When writing or reviewing asynchronous code, consult `docs/internal/Liveness.md` for guidance. ## Logging & Error Handling - Elasticsearch should prefer its own logger `org.elasticsearch.logging.LogManager

grafanaAGENTS.md

grafana

grafana/grafana/AGENTS.md

security issue reporting page](https://grafana.com/legal/report-a-security-issue/) and not directly in this repository. ## Code Review Rules ### Metrics - Review metric changes against `contribute/backend/instrumentation.md`, especially registration, duplicate registration, label cardinality

No other project uses it yet

OpenAIReads credentialsAGENTS.md

openai-cookbook

openai/openai-cookbook/AGENTS.md

article, Markdown or MDX file, and notebook Markdown cell. For editorial notebook changes, preserve code cells, outputs, execution counts, attachments, cell order, and metadata. Fix safe issues, rerun the relevant

No other project uses it yet

colbymchenryDeletes or force-pushesAGENTS.md

codegraph

colbymchenry/codegraph/AGENTS.md

summing per-turn assistant usage**, not `result.usage` (last-turn only in current Claude Code). See `docs/benchmarks/call-sequence-analysis.md`.) The mechanism that drives everything here: **an agent falls back to Read/Grep the instant

No other project uses it yet