weave-help
Skill5 starsChanged 2 months agoMITView on GitHub
Quick-reference card for Weave levels, skills, CLI commands, and how to call them from Claude Code, Codex, Gemini/Antigravity, and OpenCode. Use when the user says "weave help", "/weave-help", "what weave commands", or "how do I use weave". One-shot display, changes nothing.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/weave.mdcCursor rule
- weave-auditSkill
- weave-debtSkill
- weave-gainSkill
- weave-planSkill
- weave-promptSkill
- weave-reviewSkill
- weaveSkill
The file
Files
Weave Help
Display this card. One-shot: do not change the mode or write files.
Levels
| Level | Switch | What changes |
| lite | /weave lite | Build what is asked; name the smaller alternative in one line. |
| full | /weave full | Contract, plan for complex work, reuse ladder, verification, fidelity check. Default. |
| ultra | /weave ultra | Challenge the requirement first; deletion before addition. |
| off | /weave off, "stop weave", "normal mode" | Disabled. |
The level is saved per project in .weave/mode.
Skills
| Skill | What it does |
| weave | The workflow itself: understand, contract, plan, build the minimum, verify. |
| weave-plan | Contract and step plan saved to .weave/state.md. No edits. |
| weave-review | Current diff: what to cut, reuse, or revert (L42: reuse: ... lib/x.js). |
| weave-audit | Whole repo: ranked list of what to delete. |
| weave-debt | Ledger of weave: markers and open state items. |
| weave-gain | Measured output and token savings for this project. |
| weave-prompt | Plain-text task brief for Claude, Gemini, or Codex. |
| weave-help | This card. |
How to call them
| Agent | Form |
| Claude Code | /weave-plan, /weave-review, ... |
| Codex CLI | $weave-plan, $weave-review, ... |
| Gemini / Antigravity | /weave-plan, /weave-review, ... (from commands/*.toml) |
| OpenCode | /weave-plan, /weave-review, ... (from .opencode/commands/) |
CLI
| Command | What it does |
weave doctor | Check the installation. |
weave mode [level] | Show or set the level. |
weave prompt [claude|gemini|codex] "task" | Print a task brief; with no task, uses .weave/state.md. |
weave gain / weave discover | Measured savings / missed savings in past sessions. |
weave recall <id> | Full original output of a condensed or failed run. |
weave memory save|load|list|show|delete <name> | Snapshots of the working memory. |
weave exec -- <command> | Run a command through the output filter. |
Failed commands, file reads, and explicit JSON output are always shown in full.
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.