skills/add-git-tag/SKILL.md
Create annotated Git milestone tags. Use when completing a phase, releasing a version, marking a research checkpoint, or generating a milestone summary from git history before tagging.
npx skillsauth add a-green-hand-jack/ml-research-skills add-git-tagInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this workflow whenever the user wants to mark a milestone in the current Git repository with a tag. The agent may use a read-only sidecar to draft milestone content from git history, but the main agent must keep the human gates before creating or pushing a tag.
gpt-5.3-codex-spark through codex exec --ephemeral.read-only.git tag -a; before git push origin <tag>..agent/sidecars/<task-id>/prompt.md, output.md, model.json, and decision.md.If the user already supplied all three items, use them directly:
v0.1.3)If one or more items are missing, do not force the user to fill the whole form from scratch. First inspect the repo state and, when available, use sidecar-task-runner to draft a proposal from recent commits, tags, README or changelog snippets, and project memory.
Read-only context command:
REPO=$(git rev-parse --show-toplevel 2>/dev/null) && \
git -C "$REPO" log --oneline --decorate -30 && \
git -C "$REPO" status --short && \
git -C "$REPO" tag --sort=-version:refname | head -10
Suggested sidecar prompt:
Use recent git history and visible project docs to propose an annotated milestone tag.
Do not create a tag, commit, push, or edit production files.
Return:
- suggested tag name
- achievements as concise bullets with commit evidence
- next plans as concise bullets inferred from docs or TODOs
- uncertainty notes where the evidence is weak
If sidecar execution is not available, use the read-only git context yourself and ask the user only for the missing or uncertain fields.
After collecting user input or a sidecar proposal, compose and display the full annotated tag message for the user to review. The format should be:
Tag: <version>
Date: <current date, YYYY-MM-DD>
## ✅ This Phase — What Was Achieved
- <bullet 1>
- <bullet 2>
...
## 🚀 Next Phase — What's Planned
- <bullet 1>
- <bullet 2>
...
Ask the user: "Does this look good? Should I go ahead and create the tag?"
Wait for confirmation before proceeding.
// turbo Detect the current Git project root and check status:
REPO=$(git rev-parse --show-toplevel 2>/dev/null) && \
git -C "$REPO" log --oneline -5 && \
git -C "$REPO" status --short && \
git -C "$REPO" tag --sort=-version:refname | head -5
This gives the last 5 commits, any uncommitted changes, and the 5 most recent existing tags.
// turbo Only after user confirmation, detect the repo root and create the tag:
REPO=$(git rev-parse --show-toplevel)
git -C "$REPO" tag -a "<version>" -m "<full_tag_message>"
Where:
<version> is the tag name from the user (e.g. v0.1.3)<full_tag_message> is the full formatted message from Step 2Ask the user: "Would you like to push the tag <version> to the remote repository (origin)?"
If yes: // turbo
REPO=$(git rev-parse --show-toplevel)
git -C "$REPO" push origin "<version>"
If no, inform the user that the tag was created locally and can be pushed later with:
git push origin <version>
REPO=$(git rev-parse --show-toplevel)
git -C "$REPO" tag -n1 <version>
Report the outcome to the user:
git tag -n1 --sort=-version:refnametesting
Bootstrap project-local ml-research-skills. Use from global installs when creating a new ML research project, enabling this collection in an existing ML research repo, or deciding whether to install the full bundle locally. Route to project-init for new projects; do not handle paper or experiment work directly.
development
Route project operations tasks — git, memory, bootstrap, remote, workspace, code review, timeline, ops — to the correct skill. Use when the task involves commits, pushes, worktrees, project memory, enabling project-local skills, SSH/server coordination, sidecar runners, or audits. Do not solve the ops task directly.
testing
Route ML/AI paper writing tasks to the correct skill — contract planning, prose drafting, section writing, consistency editing, review simulation, rebuttal, submission, or citation work. Use when the task involves writing, revising, reviewing, or submitting a paper instead of guessing between paper-writing-assistant, paper-writing-contract-planner, paper-reviewer-simulator, auto-paper-improvement-loop, or citation skills. Do not draft prose directly.
data-ai
Project-local router for ML research skill selection. Use inside an initialized ML research project, or while maintaining this skill repo, when the user describes an ML research/paper/experiment/discovery/ops/release workflow and may not know the skill; route to a domain router or high-signal leaf. Do not use for generic non-ML projects.