Borrowing it
Nothing to install: this file belongs to vladm3105/aidoc-flow-framework. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/vladm3105/aidoc-flow-framework/dev/.agents/skills/git-techniques/SKILL.mdgit clone --depth 1 https://github.com/vladm3105/aidoc-flow-frameworkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/vladm3105/aidoc-flow-framework/git-techniques)<a href="https://agentmods.dev/skills/vladm3105/aidoc-flow-framework/git-techniques"><img src="https://agentmods.dev/badge/skills/vladm3105/aidoc-flow-framework/git-techniques/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vladm3105/aidoc-flow-framework/git-techniques"><img src="https://agentmods.dev/badge/skills/vladm3105/aidoc-flow-framework/git-techniques.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00026 | $0.00905 |
| Opus 5.5 | $0.00010 | $0.00362 |
| Sonnet 5.5 | $0.00005 | $0.00181 |
| Haiku 4.5 | $0.00003 | $0.00090 |
Grade A, and why
git-techniques scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Git Techniques
Use for advanced Git investigation, recovery, and debugging: reflog recovery, history search, line history, blame follow-up, bisect, fixup/autosquash, worktrees, stash inspection, and large-repo checkout options. Routine commits, pushes, branches, and PRs follow the bundled git safety skill and the repository's AGENTS.md; repository worktree, branch, and merge rules win over anything below.
Safety contract
- Default to read-only commands:
status,log,show,diff,blame,reflog,log -S,log -G,log -L, andworktree list. - Do not run a mutating operation unless the user explicitly requested that exact operation. This includes
reset,rebase,commit --amend,pushin any force form,worktree add/remove, sparse-checkout or clone-filter changes, broad stash operations, global/system config changes,maintenance, and blame-ignore-revs changes. - Do not run interactive commands such as
add -p,checkout -p,stash -p, orrebase -iin automation unless the user explicitly asks and provides an interactive context. Prefer named paths and explicit non-interactive commands. - Preserve unrelated working-tree changes. Name the files being staged, inspected, or recovered. Never commit or push changes the session did not make or cannot explain.
- Before an explicitly authorized discard, use the bundled
gitrecovery-save path where applicable. - Never rewrite pushed or shared history by default. If a remote rewrite is explicitly authorized and repository policy permits it, prefer
--force-with-leaseover--force.
Techniques
- Recovery: use
git reflogto find the priorHEAD@{n}, inspect it withgit show HEAD@{n}, and only then perform an explicitly approved reset. Reflog entries are normally retained about 90 days, not guaranteed. - History search: use
git log -S "<string>" --onelinefor added/removed occurrences,git log -G "<regex>"for pattern matches, and--pickaxe-regexwhen needed. Usegit log --greponly for commit-message search. - Line history: use
git log -L <start>,<end>:<file>orgit log -L :<function>:<file>to review diffs for specific lines or functions. - Blame follow-up: use
git blame -w -C -M <file>to look past whitespace, moves, and copies. Changing blame-ignore configuration or committing an ignore-revs file needs explicit approval. - Stash inspection: prefer named stashes, inspect with
git stash listandgit stash show -p stash@{n}, and apply rather than drop when uncertain. Include untracked files or broad paths only on explicit request. - Fixup/autosquash: use
git commit --fixup=<commit>plus autosquash rebase only for explicitly requested local history rewrites. Do not reorder or squash shared history without explicit approval. - Bisect: require a clean tree, an explicitly approved test command with trustworthy exit codes, and always finish with
git bisect reset. Report the culprit commit and the verification evidence. - Worktrees: defer to the repository's worktree flow. Do not create second clones or stray worktrees to avoid stashing; remove only worktrees the task created, after confirming no needed work remains.
- Partial staging: split commits by explicit paths rather than interactive hunk editing in automation.
- Large repositories: use shallow, blob-filter, or sparse-checkout clones only for new checkouts the user explicitly requested. Never initialize or restructure cloning inside an existing source-controlled tree.
- Performance and display settings: propose aliases and config changes for the user to approve; do not write global config without an explicit request.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 36 lines · 26 tokens per session scan A cc56e957d4d8
git-techniques is a skill published in the GitHub repository vladm3105/aidoc-flow-framework (17 stars, last pushed yesterday), licensed MIT. It adds 26 tokens to every session and 905 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-29.
Other skills, from other repositories
agent-merge-conflict-arbiter
Neutral arbiter for merge conflicts between two agents.
git-advanced-workflows
Master advanced Git workflows including rebasing, cherry-picking, bisect, worktrees, and reflog to maintain clean history and recover from any situation. Use when managing complex Git histories, collaborating on feature branches, or troubleshooting repository issues.
enable-check-includes-strict
Enable and migrate Chromium targets and directories to checkincludesstrict = true, diagnosing and resolving header include violations as part of the checkincludesstrict LSC. Use when adding directories to the allowlist, enabling strict header checks on targets, or resolving CHECKINCLUDESSTRICTUNTRIAGED targets.
resolve-handoff
Emit the complete Resolve handoff with exact modes, outcomes, test evidence, and publication state.
reproduce-and-fix-issues
Reproduce triaged Slack bugs through a configured app-control adapter, verify existing fixes, and open a bounded draft pull request only after before-and-after proof. Use only from the configured Benny repro automation.
cy-orquestrate-issue
Orchestrate GitHub bug resolution through a new CompozyOS managed worktree and Claude Opus 5.5 Medium session per issue. Investigate reported bugs, create missing issues, and dispatch an executor responsible for a detailed PR, green CI, and all CodeRabbit and Greptile feedback. Use when asked to delegate issue…