my-skills / zcode
AGENTS.md0 starsChanged 9 days agoMITView on GitHub
ZCode has no separate rules files — all rules live in this AGENTS.md, which is loaded automatically at session start. Skills live in .zcode/skills/ and load on demand. Always answer in English — even when the user writes or asks in another language, or explicitly requests a reply in that language. Work and information gathering always happen in English. Skills live in .zcode/skills/ (31 skills, standard SKILL.md format). See README.md for the "what it does / why it was created"…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 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.
In this repository34 other files it gives its agents.
- chatgpt/AGENTS.mdAGENTS.md
- opencode/AGENTS.mdAGENTS.md
- claude/CLAUDE.mdCLAUDE.md
- agent-doc-authoringSkill
- agent-experienceSkill
- ai-image-promptsSkill
- api-behavior-flowSkill
- architecture-and-performanceSkill
- behavioral-testingSkill
- behavior-auditSkill
- browser-tabs-uiSkill
- code-qualitySkill
- context-compactSkill
- design-principlesSkill
- design-to-deliverySkill
- design-validationSkill
- e2e-test-authoringSkill
- engineering-reviewSkill
- excalidrawSkill
- external-ai-consultSkill
- google-omni-videoSkill
- llm-agent-tool-disciplineSkill
- parallel-subagentsSkill
- qa-test-scenariosSkill
- reliable-executionSkill
- research-and-reuseSkill
- server-security-hardeningSkill
- session-handoffSkill
- shadcn-librarySkill
- sudo-handoffSkill
- task-intakeSkill
- tauri-v2-docsSkill
- test-failure-triageSkill
- user-behavior-flowSkill
The file
We split this AGENTS.md into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 7
Always answer in English — even when the user writes or asks in another language, or explicitly requests a reply in that language.
- Distinctive: line 12
Never commit, push, or create a PR without explicit user approval.
- Very specific: line 19
Before any operation that can cause irreversible loss of data, state, configuration, or system behavior (migration, schema change, data …
- Distinctive: line 37
File size: keep files under ~1000 lines.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
AGENTS.md — ZCode
ZCode has no separate rules files — all rules live in this AGENTS.md, which is loaded automatically at session start. Skills live in .zcode/skills/ and load on demand.
Language (Always English)
Always answer in English — even when the user writes or asks in another language, or explicitly requests a reply in that language. Work and information gathering always happen in English.
Safety Rules
Git safety
Never commit, push, or create a PR without explicit user approval.
- Explicit means the user literally says "commit this", "push it", or "submit a PR".
- Not explicit: "go ahead", "do it", "apply it", "fix it", "sounds good", "yes".
- Showing diffs, staging, and suggesting commit messages are always OK. When uncertain, ask.
Irreversible operations
Before any operation that can cause irreversible loss of data, state, configuration, or system behavior (migration, schema change, data manipulation, rm/drop/delete, config edits), ensure a backup exists or the change is reversible. Prefer additive, reversible changes.
Agent Behavior
- Inspect before editing — never modify a file you have not read.
- Do not assume requirements — when ambiguity affects architecture or behavior, ask before coding.
- Smallest change that satisfies the requirement; preserve existing behavior; avoid unrelated refactoring.
- Match existing patterns over inventing new ones; prefer fewer dependencies.
- Verify before uncertain changes — check official documentation or source instead of guessing.
- Report honestly — never present unverified work as done; say plainly when something is partial or failed.
Engineering Preferences
- Commits: short, meaningful Conventional Commits —
<type>: <short description>withfeat,fix,revert,chore,docs. Keep commits small, focused, and atomic. - Naming: names describe actual behavior — avoid
handle(),process(),doSomething(); preferfetchUserProfile(),validatePaymentToken(). - Explicit over implicit: make contracts visible, name things clearly, define data shapes, avoid hidden behavior.
- Simple over clever: simple solutions over flexible ones; maintainability over cleverness; fewer dependencies.
- No speculative features (YAGNI) — build only what is needed now.
- File size: keep files under ~1000 lines; split into focused modules when approaching the limit.
Available Skills
Skills live in .zcode/skills/ (31 skills, standard SKILL.md format). See README.md for the "what it does / why it was created" table.
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.

