The full playbook · v0/plugins/11ai-operator-git/skills/11ai-operator-git-faq/SKILL.md
11ai Git operator FAQ
Answer questions about this plugin by routing them to the plugin's own files, not from
memory. This skill owns no behavior: the sibling skills' contracts and references are the
truth, and the routing table below says which file answers which question. The only
content this skill owns is the job-family map and the glossary.
Mind the boundary with the cheatsheet: questions about Git itself (which command, which
flag) belong to the cheatsheet skill; questions about how these skills behave (what they
promise, guard, refuse, or report) belong here.
Answer contract
- Read the routed source before answering. Never answer a behavior question from memory.
- Cite the file, and the section when one is named, in every answer.
- If the sources do not cover the question, say so and name the skill that owns the
behavior. Do not guess.
- Escalate tiers when the question demands it: a question about what a skill promises
stops at its contract; coverage and lookup questions go to the references.
- If a routed sibling file is missing because only this skill was installed, read the
same skill's page at https://ai.rj11.io/skills/ followed by the skill name, or the
rj11io/11ai repository on GitHub, instead.
Ground-truth ladder
| Tier |
Meaning |
Sources |
| contract |
What a skill promises |
Sibling SKILL.md files |
| reference |
Lookup tables and matrices |
Files under sibling references directories |
| behavior |
What actually happens, asserted |
Sibling scripts and tests |
This plugin ships no scripts or tests, so no routing row carries the behavior tier; the
contracts and references are the deepest available sources.
Covered skills
11ai-operator-git-setup — machine configuration, identity, signing, and authentication.11ai-operator-git-integrations — wiring a repository to remotes, CI, LFS, and provider tooling.11ai-operator-git-status — read-only inspection of repository state.11ai-operator-git-stage — staging and unstaging only the intended changes.11ai-operator-git-commit — creating one reviewed local commit.11ai-operator-git-conventional-commits — composing and validating a Conventional Commits message.11ai-operator-git-sync — fetching, pulling, and pushing with safeguards.11ai-operator-git-branches — listing, creating, switching, renaming, and deleting branches.11ai-operator-git-worktrees — checking out two branches at once.11ai-operator-git-tags-and-releases — creating, publishing, and correcting tags.11ai-operator-git-stash — shelving and restoring work deliberately.11ai-operator-git-submodules — adding, pinning, updating, and removing submodules.11ai-operator-git-hooks — writing, sharing, debugging, and bypassing local hooks.11ai-operator-git-pull-requests — opening, reviewing, and landing pull requests.11ai-operator-git-bisect — finding the commit that introduced a regression.11ai-operator-git-recovery — undoing and recovering work within an explicit risk boundary.11ai-operator-git-cheatsheet — quick command, flag, and what-should-I-use answers.11ai-operator-git-troubleshooting — diagnosing failures from read-only evidence.
Which skill do I want?
Grouped by job; the plugin README's Choose a skill table carries the full per-skill
detail, and the cheatsheet's decision guide picks a command when the job is Git itself.
| Job |
Skills |
| Daily flow |
status, stage, commit, conventional-commits |
| Sharing work |
sync, pull-requests, integrations |
| Parallel work |
branches, worktrees, stash |
| Versions and history |
tags-and-releases, submodules, bisect |
| Automation |
hooks |
| Safety nets |
recovery, troubleshooting |
| Learning and setup |
cheatsheet, setup |
Routing table
Every row: a question, the file that answers it, the text to find in that file, and its
tier on the ground-truth ladder.
Choosing a skill
| Question |
Source |
Anchor |
Tier |
| Which Git skill fits my task? |
../../README.md |
"## Choose a skill" |
reference |
| Where do I get a quick command or flag answer without a full workflow? |
../11ai-operator-git-cheatsheet/SKILL.md |
"## Quick decision guide" |
contract |
| Which command does a given job, and what are its safety notes? |
../11ai-operator-git-cheatsheet/references/command-matrix.md |
"# Git command matrix" |
reference |
Inspection and daily flow
| Question |
Source |
Anchor |
Tier |
| How do the skills inspect repository state without changing it? |
../11ai-operator-git-status/SKILL.md |
"## Inspect in layers" |
contract |
| How is staging kept to only the intended paths or hunks? |
../11ai-operator-git-stage/SKILL.md |
"## Stage deliberately" |
contract |
| What must be true before a commit is created? |
../11ai-operator-git-commit/SKILL.md |
"## Pre-commit gate" |
contract |
| How is a Conventional Commits message composed and validated? |
../11ai-operator-git-conventional-commits/SKILL.md |
"## Validate before committing" |
contract |
Sharing work
| Question |
Source |
Anchor |
Tier |
| What safeguards apply to fetching, pulling, and pushing? |
../11ai-operator-git-sync/SKILL.md |
"## Push deliberately" |
contract |
| When is a force push allowed, and in what form? |
../11ai-operator-git-sync/SKILL.md |
"force-with-lease" |
contract |
| How is a pull request opened, kept current, and landed? |
../11ai-operator-git-pull-requests/SKILL.md |
"## Land it" |
contract |
| How is a repository wired to remotes, CI, LFS, or provider tooling? |
../11ai-operator-git-integrations/SKILL.md |
"## Name the seam" |
contract |
| Which integration seams exist and how is each verified end to end? |
../11ai-operator-git-integrations/references/integrations.md |
"# Git integrations reference" |
reference |
Branching and parallel work
| Question |
Source |
Anchor |
Tier |
| How are branches created, switched, renamed, and deleted safely? |
../11ai-operator-git-branches/SKILL.md |
"## Common operations" |
contract |
| How do I work on two branches at the same time? |
../11ai-operator-git-worktrees/SKILL.md |
"## Add a worktree" |
contract |
| When can a worktree be removed, and which removals need approval? |
../11ai-operator-git-worktrees/SKILL.md |
"## Remove and prune" |
contract |
| How is work shelved and restored, and when are stashes deleted? |
../11ai-operator-git-stash/SKILL.md |
"## Delete only by request" |
contract |
History, versions, and automation
| Question |
Source |
Anchor |
Tier |
| How are tags created, published, and corrected after they are pushed? |
../11ai-operator-git-tags-and-releases/SKILL.md |
"## Correct a tag" |
contract |
| How are submodules added, pinned, updated, and removed completely? |
../11ai-operator-git-submodules/SKILL.md |
"## Remove one completely" |
contract |
| How are hooks written so the team keeps them, and how are they shared? |
../11ai-operator-git-hooks/SKILL.md |
"## Write a hook that people keep" |
contract |
| Can a local hook be bypassed, and what does that mean for enforcement? |
../11ai-operator-git-hooks/SKILL.md |
"no-verify" |
contract |
| How is a regression hunted down to a single commit? |
../11ai-operator-git-bisect/SKILL.md |
"## Establish a reliable check first" |
contract |
| How does a bisect session end without leaving the repository detached? |
../11ai-operator-git-bisect/SKILL.md |
"## Recover and reset" |
contract |
Recovery and safety
| Question |
Source |
Anchor |
Tier |
| How is lost or broken work recovered, and what evidence comes first? |
../11ai-operator-git-recovery/SKILL.md |
"## Capture evidence" |
contract |
| How does recovery pick the least destructive operation? |
../11ai-operator-git-recovery/SKILL.md |
"## Choose the smallest operation" |
contract |
| Which actions always require explicit approval across every skill? |
../../README.md |
"## Shared safety contract" |
reference |
| How are identity, signing, and authentication set up on a new machine? |
../11ai-operator-git-setup/SKILL.md |
"## Configure authentication and signing" |
contract |
Troubleshooting
| Question |
Source |
Anchor |
Tier |
| How is a Git failure diagnosed before anything is changed? |
../11ai-operator-git-troubleshooting/SKILL.md |
"## Classify before fixing" |
contract |
| Where is the symptom-to-cause triage table? |
../11ai-operator-git-troubleshooting/references/triage-matrix.md |
"# Git troubleshooting triage matrix" |
reference |
Glossary
Definitions only; the commands, flags, and defaults live in the routed sources above.
| Term |
Meaning |
| Shared safety contract |
The plugin-wide rules every skill follows about inspection, approval, and what is never done |
| Path-scoped command |
A command limited to named files or folders so it cannot touch unrelated work |
| Read-only inspection |
Looking at repository state with commands that change nothing; the default first step |
| Fast-forward |
Moving a branch pointer ahead without creating a merge, possible only when histories have not diverged |
| Force push safeguard |
The rule that rewriting a published branch needs explicit approval and the safer lease-checked form |
| Seam |
A named connection point between the repository and an outside system, such as a remote or CI |
| Pinned commit |
The exact submodule commit a parent repository records and expects |
| Reflog |
The local journal of where branch tips pointed before each change; the main recovery source |
| Server-side check |
An enforcement that runs on the host, which a local hook bypass cannot skip |
| Triage matrix |
The lookup table mapping failure symptoms to likely causes and the smallest repair |