cladding-init
Use only when the user explicitly names Cladding and asks to initialize, adopt, or refresh it; never use for an ordinary project…
@qwerfunchInstall SkillDevelopment
Skill
/clarify
Advance Cladding onboarding after the user answers a pending product question. Use the MCP prepare/apply flow, preserve the answer verbatim, and never invent answers. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore
Claude Code
$ npx -y skills add purpleeddy/cladding --skill clarify --agent claude-code
Codex
$ npx skills add purpleeddy/cladding --skill clarify -a codex -y
Cursor
$ npx skills add purpleeddy/cladding --skill clarify -a cursor -y
Gemini CLI
$ npx skills add purpleeddy/cladding --skill clarify -a gemini-cli -y
GitHub Copilot
$ npx skills add purpleeddy/cladding --skill clarify -a github-copilot -y
OpenCode
$ npx skills add purpleeddy/cladding --skill clarify -a opencode -y
Windsurf
$ npx skills add purpleeddy/cladding --skill clarify -a windsurf -y
SKILL.md
clarify.SKILL.mddescription: Advance Cladding onboarding after the user answers a pending product question. Use the MCP prepare/apply flow, preserve the answer verbatim, and never invent answers. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.
Cladding clarify
Use this workflow only for an answer contained in a new user message after Cladding displayed the pending question. Never infer, synthesize, or reuse an answer from an initialization turn. If no new user answer exists, stop without calling either clarify tool.
Use this workflow only after Cladding initialization returned a pending question and the user has answered it.
1. Call `clad_prepare_clarify` with the user's answer verbatim. 2. Read the returned prompt, current state, and artifacts. Draft the structured refinement required by `clad_clarify` using the current host model. 3. Call `clad_clarify` with the same answer, the one-time token, and the draft. 4. Ask `nextQuestion` verbatim when one remains. 5. If the result is `needs_review`, show the proposal diffs for every `pendingReview` target and wait for explicit user approval. Only then call `clad_resolve_onboarding_review` with the approved targets. Never describe onboarding as complete while review remains. 6. Report completion only when the returned status is `done`.
Do not run `clad clarify` in a shell from an AI host. Do not use MCP sampling. Never answer a product question on the user's behalf. A stale, malformed, replayed, or answer-mismatched apply request is a no-op and must be prepared again.
Read more
description: Advance Cladding onboarding after the user answers a pending product question. Use the MCP prepare/apply flow, preserve the answer verbatim, and never invent answers. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.
Cladding clarify
Use this workflow only for an answer contained in a new user message after Cladding displayed the pending question. Never infer, synthesize, or reuse an answer from an initialization turn. If no new user answer exists, stop without calling either clarify tool.
Use this workflow only after Cladding initialization returned a pending question and the user has answered it.
1. Call `clad_prepare_clarify` with the user's answer verbatim. 2. Read the returned prompt, current state, and artifacts. Draft the structured refinement required by `clad_clarify` using the current host model. 3. Call `clad_clarify` with the same answer, the one-time token, and the draft. 4. Ask `nextQuestion` verbatim when one remains. 5. If the result is `needs_review`, show the proposal diffs for every `pendingReview` target and wait for explicit user approval. Only then call `clad_resolve_onboarding_review` with the approved targets. Never describe onboarding as complete while review remains. 6. Report completion only when the returned status is `done`.
Do not run `clad clarify` in a shell from an AI host. Do not use MCP sampling. Never answer a product question on the user's behalf. A stale, malformed, replayed, or answer-mismatched apply request is a no-op and must be prepared again.
Stats
16
Stars
4
Forks
Maintained
Maintenance
TypeScript
Language
MIT
License
Last commit
Created
Ships withcladding
For an organization to trust AI with its code, three things must hold — trust, traceability, and stability at scale. cladding wraps your AI coding agent: your intent goes in before it writes, and the result is verified against your spec after, so those three are earned, not assumed. First L4 implementation of the Ironclad standard.
1619 skills6 agents1 MCP
Stars on GitHub
Momentum
- stars in total
- 16
Other skills on cladding.
@qwerfunchInstall Skillblind-author
Impl-blind test/oracle author — writes conformance tests from a spec-only brief. Tool-restricted by definition (no…
@qwerfunchInstall Skillchangelog
Render release notes / a changelog from the cladding spec. Use when the user asks for release notes, a changelog, 릴리즈 노트, 변경 이력,…
@qwerfunchInstall Skillcheck
Run every Iron Law stage and the drift detector suite. Use when the user wants the full project health snapshot, a CI gate, or to…
@qwerfunchInstall Skillcheckpoint
Record a feature checkpoint event pinning the current git HEAD plus a spec digest, so a later rollback can restore the exact…
@qwerfunchInstall Skilldeveloper
Implementer — writes production code, tests, and migrations. The "generic engineer" fallback when no narrower specialist exists.…
@qwerfunchInstall SkillMore skills from other plugins.
Auto-invoked
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…
@obraInstall Skillask-matt
Ask which skill or flow fits your situation. A router over the skills in this repo.
@mattpocockInstall SkillAuto-invoked
Questions about clarify.
- Which coding agents can use clarify?
- clarify installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.