git-bisect SKILL.md skillA

git-bisect from product-on-purpose/agent-skills-toolkit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add product-on-purpose/agent-skills-toolkit/git-bisect@v0.1.0

Pins v0.1.0 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
product-on-purpose/agent-skills-toolkit
Path
tests/fixtures/audit-corpus-labelled/02-excellent-orphan/skills/git-bisect/SKILL.md
Versions recorded
1
Latest version
v0.1.0, committed Nov 1, 2026
Size
2,290 bytes
Repository stars
3
License
Apache-2.0

Finds the commit that introduced a regression by driving git bisect with an automated test command and reporting the culprit with its diff. Use when a test that used to pass now fails, when the user asks which commit broke something, or mentions bisecting, regression hunting, or 'when did this start'.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# git-bisect

## Purpose

Find the first bad commit between a known-good and a known-bad ref by letting `git bisect run` drive the search with a test command, then report the culprit commit with its diff and the reasoning.

## Inputs to establish first

1. A `good` ref where the behavior was correct (a tag, a sha, or "the last release"). If the user does not know one, propose `git log --oneline -30` and ask them to pick the oldest commit they trust.
2. A `bad` ref (default `HEAD`).
3. A command that exits 0 when the behavior is correct and non-zero when it is broken. Prefer the narrowest test: one test file, one test name. If none exists, write a throwaway script under `/tmp` that reproduces the report and exits by its outcome, and show it to the user before running.

## Procedure

1. Confirm the working tree is clean (`git status --porcelain`); stash or refuse if it is not, because bisect checks out commits.
2. Sanity-check the endpoints before bisecting: run the command at `bad` (must fail) and at `good` (must pass). If either check does not behave, stop and say so; bisecting on a flaky or mis-specified command wastes every step.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add product-on-purpose/agent-skills-toolkit/git-bisect@sha256:dbb4a9203a8f7d72

The line at the top of this page pins the label v0.1.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_4olfpto6pogptyvb.svg)](https://markdownregistry.com/a/art_4olfpto6pogptyvb)

0 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Nov 1, 2026.

versioncommittedcommitsizeaudit
v0.1.0 latest2026-11-01 6dd7f43 2,290 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (2290 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

product-on-purpose/agent-skills-toolkit · 3 stars · license Apache-2.0 · pushed 2026-11-08 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_4olfpto6pogptyvb
GET https://markdownregistry.com/api/v1/resolve?ref=product-on-purpose/agent-skills-toolkit/git-bisect
GET https://markdownregistry.com/api/v1/blob/dbb4a9203a8f7d726b37ccc025bae6a944bb6ac90c2cd4c1529610dc84f8ac8f

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from product-on-purpose/agent-skills-toolkit

AGENTS.md AGENTS.md file
product-on-purpose/agent-skills-toolkit · AGENTS.md
git:20261024.a6e6107 · audit A · 3 stars
DESIGN.md@docs/internal DESIGN.md file
product-on-purpose/agent-skills-toolkit · docs/internal/DESIGN.md
git:20260528.c79be5c · audit A · 3 stars
DESIGN.md@docs/internal/research DESIGN.md file
product-on-purpose/agent-skills-toolkit · docs/internal/research/DESIGN.md
git:20260610.74583cc · audit A · 3 stars
probe-duplicate SKILL.md skill
product-on-purpose/agent-skills-toolkit · docs/internal/vendor-watch/probes/components-share-one-namespace/probe-collision-a/skills/probe-duplicate/SKILL.md · PROBE FIXTURE side A. Two plugins ship a skill with this exact directory name and this exact skill name. Invoke it and…
git:20260820.9a652c6 · audit A · 3 stars
probe-duplicate SKILL.md skill
product-on-purpose/agent-skills-toolkit · docs/internal/vendor-watch/probes/components-share-one-namespace/probe-collision-b/skills/probe-duplicate/SKILL.md · PROBE FIXTURE side B. Two plugins ship a skill with this exact directory name and this exact skill name. Invoke it and…
git:20260820.9a652c6 · audit A · 3 stars
askit-backlog SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-backlog/SKILL.md · Creates, triages, and prunes a plugin's two backlogs (new-component proposals and enhancements) to the Advanced Skill…
v0.1.0 · audit A · 3 stars
askit-build-agents-md SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-build-agents-md/SKILL.md · Creates and improves a plugin's AGENTS.md (the agent navigation and instructions entrypoint) to the Advanced Skill…
v0.1.0 · audit A · 3 stars
askit-build-chain-contract SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-build-chain-contract/SKILL.md · Creates and improves a plugin's chain contract (agents/_chain-permitted.yaml) to the Advanced Skill Library Standard…
v0.1.0 · audit A · 3 stars
askit-build-command SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-build-command/SKILL.md · Creates and improves Claude slash commands (commands/<name>.md) that map to a skill, to the Advanced Skill Library…
v0.1.0 · audit A · 3 stars
askit-build-docs SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-build-docs/SKILL.md · Creates and improves a plugin's documentation across modes (readme, quickstart, tutorial, how-to, reference, glossary…
v0.2.1 · audit A · 3 stars
askit-build-hook SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-build-hook/SKILL.md · Creates and improves event-driven hooks (Advanced tier) for a plugin to the Advanced Skill Library Standard. Use when…
v0.1.0 · audit A · 3 stars
askit-build-mcp SKILL.md skill
product-on-purpose/agent-skills-toolkit · skills/askit-build-mcp/SKILL.md · Creates and improves MCP server definitions (a portable .mcp.json) for a plugin to the Advanced Skill Library Standard…
v0.1.0 · audit A · 3 stars

Every file in product-on-purpose/agent-skills-toolkit

Browse by kind, by grade A, or by owner.