Stats
Actions
Available In
Tags
What's Inside
Agents2
spec-maker
/spec-maker
Use this agent to develop a feature from a specification file or detailed requirements document. Transforms written specifications into production-ready code with deep analysis and knowledge gathering. Example:\n\n<example>\nuser: "I have a spec for a new API endpoint in specs/user-authentication.md. Can you implement it?"\nassistant: "I'll use the Task tool to launch the spec-maker agent to develop this feature from the specification."\n<commentary>The user has provided a specification file that needs to be developed into code, the primary use case for this agent.</commentary>\n</example>
spec-writer
/spec-writer
Use this agent when the user requests documentation of features, architecture decisions, or implementation plans that need to be formalized into functional specifications. Trigger when the user asks for a functional spec, when a new feature needs to be designed and documented before implementation, or when complex requirements need to be clarified and structured for another developer or agent.\n\n<example>\nContext: User wants to create a new data connector and needs clear specifications before implementation.\nuser: "I need to build a connector for Salesforce that syncs contacts and opportunities"\nassistant: "I'll use the Task tool to launch the spec-writer agent to create comprehensive specifications for this connector."\n<commentary>This is a new feature requiring structured documentation and analysis of existing patterns — spec-writer should gather requirements and produce the specification.</commentary>\n</example>
Skills8
commit-writer
/commit-writer
Generate git commit messages from staged or unstaged changes. Use this skill whenever the user invokes "/rocket:commit-writer", asks for a commit message, wants to describe their git changes, says "what should I commit?", "write a commit message", "redige un message de commit", "propose un commit", or any similar request involving summarizing git diffs into a commit message.
context-clear
/context-clear
Wipe the project's rocket-managed contextualization files (lexicon, logs, locks) under .roc/rocket/. Use this skill whenever the user invokes "/rocket:context-clear", says "clear context", "reset the lexicon", "wipe the lexicon", "clean context files", "nettoie le contexte", "supprime les fichiers de contextualisation", "réinitialise le lexique", "vide le lexique", "remets le contexte à zéro", or any similar request to discard the rocket contextualization state for the current project.
context-update
/context-update
Update the project's semantic lexicon at .roc/rocket/lexicon.md from the current conversation. Use this skill whenever the user invokes "/rocket:context-update", asks to "update context", "refresh lexicon", "update the lexicon", "rebuild project vocabulary", "mets a jour le contexte", "mets a jour le lexique", "rafraichis le lexique", or any similar request. Auto-invoke when a major semantic shift just happened in the conversation (new domain concept, new architectural decision, redefinition of an existing term) and the lexicon should capture it before the next turn.
myself
/myself
Produce a precise change plan (file:line + short prose + why) for a human developer to apply themselves, instead of editing files. Manual-only — invoked by the user via "/rocket:myself" or "/myself".
no-code
/no-code
Force a chat-only response with no code writing, no file edits, no tool use that modifies the filesystem. Manual-only — invoked by the user via "/rocket:no-code" or "/no-code".
Stats
Version1.2.0
LanguagePython
Stars0
MaintenanceExcellent
LicenseMIT
Last Commit
Added
Available In
Safety Signals
Caution
Uses power tools
Uses Bash, Write, or Edit tools
Tags
Categories
