How this skill is triggered — by the user, by Claude, or both
Slash command
/agentic-doc:agd-commitThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate a Conventional Commits message from staged changes and create the commit.
Generate a Conventional Commits message from staged changes and create the commit.
Human owns the commit. Never add Co-Authored-By footer for AI assistance. The human takes full responsibility for all commits in their repository.
First, load the conventional-commits skill to understand commit format and best practices:
Use Skill tool with skill="agd:conventional-commits"
This provides:
Check git status:
git status --short
Identify unstaged files (lines where first column is space, ?, or has changes not yet staged).
Handle staging decision:
Call the AskUserQuestion tool with these parameters:
{
"questions": [{
"question": "There are unstaged changes. Would you like me to stage them?",
"header": "Stage changes",
"options": [
{"label": "Stage all", "description": "Run 'git add -A' to stage all changes"},
{"label": "Skip staging", "description": "Commit only currently staged changes"}
]
}]
}
Wait for the tool response, then:
git add -A, then continue to step 3If no unstaged changes exist, skip this step and continue directly to step 3.
Analyze the diff to determine:
Generate commit message:
<type>[scope]: <description>Important: Do NOT add Co-Authored-By footer. The commit belongs to the human user.
Get human confirmation:
Call the AskUserQuestion tool to get approval:
{
"questions": [{
"question": "Ready to commit with this message:\n\n<commit message>\n\nProceed?",
"header": "Confirm commit",
"options": [
{"label": "Yes, commit", "description": "Create the commit with this message"},
{"label": "Edit message", "description": "Provide a custom commit message instead"}
]
}]
}
Create the commit:
git commit -m "<type>[scope]: <description>"git commit -m "<type>[scope]: <description>" -m "<body line 1>" -m "<body line 2>"git commitProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub christophe1997/agent-extentions --plugin agentic-doc