commit-assisted-by
Add an `Assisted-by: <agent-name>/<model-id>` git trailer to commits made during an AI-assisted coding session. Trigger on any `git commit` activity visible in the session — either commits you make yourself, or commits the user makes outside your tool calls that surface in the conversation — even without an explicit request. Also trigger when the user explicitly asks to tag existing commits in a branch (e.g. "add assisted-by to commits in this branch").
What this skill does
# Commit Assisted-by trailer
Record the assisting agent's involvement in a commit by adding an `Assisted-by: <agent-name>/<model-id>` git
trailer. The `<agent-name>` identify the specific AI assistant (CLI/harness) and `<model-id>` is the LLM version
that contributed to the commit. This disclosure provides transparency about AI involvement in code changes.
In the command examples below, substitute `<agent-name>/<model-id>` with your own values before running anything.
## Mode 1 — commits you create yourself
When you create a commit in a session, limit the commit subject to 50 characters. Use the imperative mood in the
subject line (e.g., "Add support for X" instead of "Added support for X").
If the commit message body is needed to explain what and why, wrap it at 72 characters. Avoid verbose explanations
in the body, be as concise as possible while still conveying the necessary information.
Pass `--trailer` directly on the commit command. This is the clean path and avoids amend noise:
```
git commit -s -m "Add support for X" --trailer "Assisted-by: <agent-name>/<model-id>"
```
Note that all commits must be signed off with a real human name and email to meet the DCO requirement. Never use
your agent's name or a pseudonym in the sign-off or co-author fields.
## Mode 2 — commits that appear outside your own `git commit` invocations
Sometimes a commit lands on HEAD that you did not run yourself — the user invoked git via a shell escape (e.g.
Claude Code's `! git commit ...`) or a separate terminal. Your only window is after the fact.
Run the bundled safety-check script before amending. It prints `safe`, or `skip:<reason>` if HEAD is already
pushed, already trailered, a merge, or has a rebase/cherry-pick/revert/bisect in progress.
Invoke it with the script's full path — resolve `scripts/safety-check.sh` against the directory this SKILL.md was
loaded from, not the current working directory (the two are almost never the same):
```
<skill-dir>/scripts/safety-check.sh
```
Only when the output is exactly `safe`, amend:
```
git commit --amend --no-edit --trailer "Assisted-by: <agent-name>/<model-id>"
```
Follow up with one short line so the user knows the hash moved, e.g. "Amended HEAD to add Assisted-by trailer." If
the output starts with `skip:`, tell the user why the commit was skipped.
## Mode 3 — explicit "tag the whole branch" request
Triggered by phrases like "add assisted-by to commits in this branch", "tag the commits on this branch", or
"backfill the trailer". This is the **only** mode that rewrites more than HEAD, and only because the explicit
request authorizes it. Never enter this mode on autopilot.
### Pick the base
Default to `@{u}` — it guarantees the range is only unpushed commits, which removes the force-push question
entirely. If the branch has no upstream, ask the user for an explicit base ref rather than guessing. If the user
names a base themselves ("from the last release tag"), honor it.
### Show the plan, then confirm
1. `git log --oneline <base>..HEAD` so the user can see the commits that will be rewritten.
2. If `<base>` is anything other than `@{u}`, run `git rev-list <base>..@{u} 2>/dev/null`. If non-empty, those
commits are already on the remote — warn the user that after the rebase they will need to force-push to publish
the rewrite, and confirm before proceeding.
3. If the range contains merge commits, stop and ask how to handle them. Default rebase linearizes merges
(destructive), and `--rebase-merges` with `--exec` on merges is fragile — not worth the risk without the user's
judgment.
### Run it
The exec short-circuits on commits that already carry the trailer, so the pass is idempotent:
```
git rebase <base> --exec 'git log -1 --format=%B | git interpret-trailers --parse | grep -qi "^Assisted-by:" || git commit --amend --no-edit --trailer "Assisted-by: <agent-name>/<model-id>"'
```
Afterwards, report how many commits were tagged vs skipped. If the branch was already pushed, tell the user they
will need to run `git push --force-with-lease` themselves to publish the rewrite.
If the rebase hits a conflict, do not try to resolve it. Surface the conflict to the user and let them choose
`--continue` vs `--abort`.
## Anti-patterns
- Never run `git push` yourself in any mode, even after a rebase. Publishing rewritten history is the user's call —
surface the state and let them decide.
- Don't use `Co-authored-by:` — AI assistants are not GitHub users and the co-author convention is reserved for
human collaborators. `Assisted-by:` is the correct semantic.
- Don't fabricate or prettify either token. Use the actual agent/CLI/harness name and model ID/version.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.