shepherd-review

shepherd-review is a skill for Claude Code from mshadmanrahman/pm-pilot. It costs 78 tokens per session (1,919 once invoked), scanned A, original, MIT.

A review step for bug fixes before they are synced or submitted. It checks the active code branch, the changes made, project rules, and supporting evidence.

In plain words
What is it for?
Use it to review a bug-fix branch, check its difference from the main branch, apply project-specific checks, and flag changes that need explanation or more evidence.
Why use it?
It helps catch oversized or insufficiently supported fixes before they reach the main codebase.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents.

Part of the bug-shepherd plugin — 5 skills, 1 agent shipped together

Good fit Use it to review a bug-fix branch, check its difference from the main branch, apply project-specific checks, and flag changes that need explanation or more evidence.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/mshadmanrahman/pm-pilot/shepherd-review
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add mshadmanrahman/pm-pilot --skill shepherd-review
Clone the repo
git clone --depth 1 https://github.com/mshadmanrahman/pm-pilot

Made for: Claude Code.

Or install bug-shepherd, the plugin that ships this one along with the rest of its 5 skills, 1 agent.

Wrote 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.

agentmods badge for shepherd-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/mshadmanrahman/pm-pilot/shepherd-review/github.svg)](https://agentmods.dev/skills/mshadmanrahman/pm-pilot/shepherd-review)
Your own site
<a href="https://agentmods.dev/skills/mshadmanrahman/pm-pilot/shepherd-review"><img src="https://agentmods.dev/badge/skills/mshadmanrahman/pm-pilot/shepherd-review/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.

agentmods 80×15 button for shepherd-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/mshadmanrahman/pm-pilot/shepherd-review"><img src="https://agentmods.dev/badge/skills/mshadmanrahman/pm-pilot/shepherd-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 78 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,919 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 3 findings, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Excessive Agency · line 32
    Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
    Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
  • medium MCP Rug Pull · line 125
    npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
    Fix: Pin the version: npx @scope/[email protected]
  • medium MCP Rug Pull · line 125
    npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
    Fix: Pin the version: npx @scope/[email protected]
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00078 $0.01919
Opus 5 $0.00039 $0.00959
Sonnet 5 $0.00016 $0.00384
Haiku 4.5 $0.00008 $0.00192

Measured 4d ago against content hash e20c1542a2a2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

shepherd-review 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 4d 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.

plugins/bug-shepherd/skills/shepherd-review/SKILL.md · 205 lines

How it starts

The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/shepherd-review — Quality Gate Before Pushing

A mandatory quality check before any bug fix is pushed. Ensures minimal, focused changes with proper evidence.

Audience: Developers who fixed a bug and want to verify before creating a PR. Also useful for PMs reviewing a developer's proposed fix.

Independent review, not self-review

Run the actual review (step 3) in a fresh subagent, launched with the Agent tool, not in the session that investigated the bug and wrote the fix. That session already holds the investigation narrative and the reasoning that led to this specific diff. Asking it to also grade the diff means the same context that talked itself into the fix is now checking its own work. The audience note above already assumes a second set of eyes ("a developer... a PM reviewing a developer's fix"); this makes that true even on a solo session where the fixer and the reviewer are the same person.

Give the subagent only:

  • the diff (git diff main...HEAD)
  • the ticket ID and title, not the investigation transcript
  • review.tech_stack_rules, review.max_line_diff, review.require_evidence from .claude/triage.config.yaml
  • .claude/learning-log.md, for Check 5

It does not get the conversation that produced the fix. If a change can't be justified from the diff and the ticket alone, that's a real finding, not a gap in the subagent's context: either the diff needs a comment or it wasn't scoped as tightly as the fixer assumed.

Workflow

1. Load Configuration (current session)

Read .claude/triage.config.yaml for:

  • review.max_line_diff — flag if net lines exceed this
  • review.tech_stack_rules — project-specific checks
  • review.require_evidence — whether before/after evidence is required
  • review.allow_shell_checks: whether check: rules may run without asking

If the config does not exist, run the first-run setup in /shepherd-sync step 1 to create it from the bundled template, then continue.

2. Verify Active Session (current session)

Read the full file on GitHub · 205 lines

Changes

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.

  1. 4d ago Changed · +30 lines · +25 tokens per session e20c1542a2a2
  2. 5d ago Changed · +16 lines · +18 tokens per session ddc7ec9568bc
  3. 10d ago First seen · 159 lines · 35 tokens per session scan A 1f8655c87387

Subscribe to this mod's changes

shepherd-review is a skill published in the GitHub repository mshadmanrahman/pm-pilot (20 stars, last pushed 4d ago), licensed MIT. It adds 78 tokens to every session and 1,919 once invoked, about $0.0004 per session on Opus 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 2026-08-30.

Related

Other skills, from other repositories

verify-implementation

A workflow that runs a project’s verification skills to produce a report on coding patterns, architecture rules, and project conventions. It is intended for work after implementation, before a pull request, or during code review.

sangrokjung/claude-forge · 37 tokens

review-loop

Run the adversarial verification loop — implement, then hand the change to a fresh checker that did not write it, fix what it finds, and re-dispatch until APPROVE. Use before claiming any behavioural change is done, and on requests like "review loop", "adversarial review", "independent review", "get this verified"…

sangrokjung/claude-forge · 100 tokens

frontend-code-review

Trigger when the user requests a review of frontend files (e.g., .tsx, .ts, .js). Support both pending-change reviews and focused file reviews while applying the checklist rules.

sangrokjung/claude-forge · 42 tokens

no-mistakes

Validate committed feature-branch changes through the no-mistakes pipeline: intent, rebase, review, test, docs, lint, push, PR, and CI. Use when the user asks to run no-mistakes, ship safely, validate before pushing, or gate a change before it reaches upstream.

stevesolun/ctx · 67 tokens

ai-slop-cleaner

Post-implementation cleanup that removes AI-generated bloat while preserving functionality. Runs pass-by-pass with test verification after each pass. Activate after kraken/spark complete a feature, or when a codebase needs hygiene work.

vibeeval/vibecosystem · 50 tokens

coding-standards

Universal coding standards - naming, formatting, error handling, immutability, SOLID principles.

vibeeval/vibecosystem · 23 tokens