{"slug": "global-agents-md", "title": "Global Agents.md", "summary": "A developer published a global Agents.md file specifying autonomous coding-agent conventions, including merging loose branches into a single working tree, banning back-compat refactors and dead code, minimizing third-party dependencies, and targeting full coverage through use-case end-to-end tests. The file also prescribes per-model routing across GLM, DeepSeek and Kimi models by task type, a three-phase PR CI pipeline, and routine cleanup of caches and owned processes.", "body_md": "[Agents.md](https://gist.github.com/MelbourneDeveloper/4e3d96b98ac345d0a45d1808973e709e#file-agents-md)\n- 1 feature branch, 1 working tree: merge all loose branches/stashes/uncommitted files into it & push. Delete branches already merged into it\n- ❌ ask user questions. Decide, act, report. Act **autonomously**\n- Duplication = bug. Use [deslop](https://deslop.live/docs/for-ai/)\n- Important lints → error. Run static analysis routinely\n- 🧹 After work: delete safe system caches & unneeded build files, kill your owned processes that eat CPU or memory and check disk space\n- Refactors: violent, immediate, no back-compat. Legacy code banned → delete now\n- Minimize code; decide toward less. Delete dead code now\n- 📦 Minimize 3rd-party deps: routinely delete unused packages/crates, add none without need, consolidate overlapping deps into one, avoid those with a vulnerability history\n- Reply ≤2 sentences, emojis for emphasis. Don't stream responses\n- ✅ Tick mandatory plan checklist items as you work\n- Website docs, blogs = prose classes, CSS BUDGET 2K LOC default\n- ❌ self as git commit coauthor\n- PR CI Phases = Build/Analyze/Cache Build -> Run tests parallel with coverage (on cached build) -> Coverage checks\n- Declare all models with [typeDiagram](https://typediagram.dev/docs/language-reference.html) , generate code from them, consider[tdbin for wire protocol](https://typediagram.dev/docs/tdbin.html)\n- Specs: hierarchical non-numeric IDs, cross-ref'd across code/tests/plans/artifacts\n- Repo structure: [AgentPMO](https://www.agentpmo.dev/get-started/) . Deploy structure:[Shipwright](https://github.com/Nimblesite/Shipwright)\n- Aim 💯 coverage via use-case E2E tests. Unit tests banned unless isolation needed\n- 💰 Watch spend (GH Actions, agents, cloud: GCP/Supabase…). Minimize & prevent unneeded future spend\n- Tests: many input permutations & assertions. Prefer long tests: many user interactions, many assertions each\n- Offload simple work to pi; pick models per `/Users/christianfindlay/Documents/Code/PiNotes/MODEL-COMPARISON.md` , always each family's newest id in`pi --list-models` (today`zai/glm-5.3-flash` repetitive/fixtures/tests,`zai/glm-5.3` scoped code/reviews,`deepseek/deepseek-flash` code/oracles,`kimi-coding/k3.1-pro-highspeed` UI,`deepseek/deepseek-v4-pro` narrow audits only); bounded tasks, verify results, record model/date/results in PiNotes\n- ❌ Claude artifacts: reports/docs → markdown in repo\n- Commit after each step; don't let uncommitted files pile up\n- 🔒 Close issues only after testing the LATEST release's own binaries 📦. ❌ Never close via PRs", "url": "https://wpnews.pro/news/global-agents-md", "canonical_source": "https://gist.github.com/MelbourneDeveloper/4e3d96b98ac345d0a45d1808973e709e", "published_at": "2026-11-18 20:48:05+00:00", "updated_at": "2026-11-19 02:42:52.558513+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools"], "entities": ["Claude", "GLM", "DeepSeek", "Kimi", "GitHub Actions", "AgentPMO", "Shipwright"], "also_reported_by": [], "takeaway": "A global Agents.md that forbids asking the user questions, mandates one working tree, bans back-compat and dead code, and routes tasks to specific models (GLM-5.3-flash for fixtures, DeepSeek-flash for code, Kimi-k3.1-pro-highspeed for UI) gives coding agents a fixed autonomous operating contract.", "recommendation": {"advice": "Adopt a repo-level Agents.md that fixes autonomy rules, model routing per task type, and a build-analyze-test CI order so coding agents act without asking and commit after each step.", "playbook": "coding-agents", "playbook_url": "https://wpnews.pro/playbooks/coding-agents"}, "alternates": {"html": "https://wpnews.pro/news/global-agents-md", "markdown": "https://wpnews.pro/news/global-agents-md.md", "text": "https://wpnews.pro/news/global-agents-md.txt", "jsonld": "https://wpnews.pro/news/global-agents-md.jsonld"}}