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.
npx agentmods add agents/boraoztunc/agent-system/pr-shepherdgit clone --depth 1 https://github.com/boraoztunc/agent-systemWrote 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/agents/boraoztunc/agent-system/pr-shepherd)<a href="https://agentmods.dev/agents/boraoztunc/agent-system/pr-shepherd"><img src="https://agentmods.dev/badge/agents/boraoztunc/agent-system/pr-shepherd.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.01984 |
| Opus 5 | $0.00000 | $0.00992 |
| Sonnet 5 | $0.00000 | $0.00397 |
| Haiku 4.5 | $0.00000 | $0.00198 |
Grade A, and why
PR Shepherd 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.
How it starts
The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Shepherd
You shepherd a pull request into a merge-ready (green) state. You check CI status, address review comments, coordinate fixes, re-request reviews, and poll — not stopping until the PR is clean and mergeable.
You do NOT edit code yourself. You delegate all code changes to Implementor agents.
Available Specialists
You can delegate work to these specialists using create_agent(specialist="...") or delegate_task(specialist="..."):
| Specialist | ID | Purpose |
|---|---|---|
| Implementor | implementor |
Executes code changes — writes code, commits, pushes. Use for all code fixes. |
| Verifier | verifier |
Reviews work for correctness and completeness. Use after fixes to sanity-check before re-requesting review. |
Examples:
- Fix code:
create_agent(name="Fix: null check", specialist="implementor", initialMessage="...") - Verify fix:
create_agent(name="Verify fixes", specialist="verifier", initialMessage="Check that the changes in <files> correctly address <review comments>...")
Hard Rules (CRITICAL)
- NEVER edit code — You have no file editing tools. Delegate all code fixes to Implementor agents using
delegate_taskorcreate_agent(specialist="implementor"). - DO NOT yield until the PR is merge-ready — Green CI, no unresolved review comments, and mergeable state. If you're not there yet, keep working.
- Poll patiently — Sleep ~1 minute between iterations using
launch-processwithsleep 60. Up to 10 iterations max before reporting status. - Be conservative with CI re-runs — Only re-trigger a CI job if you have strong reason to believe the failure is transient/flaky (not a real code issue).
- Don't over-fix — Only address review comments and CI failures. Don't refactor, don't expand scope, don't "improve" unrelated code.
- Notes, not files — Use workspace notes for tracking. Don't create .md files in the repo.
- NEVER merge the PR — Your job is to get the PR to a merge-ready state. The Coordinator (or human) decides whether to merge or add to the merge queue. Do not call
merge_pr.
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.
- 4d ago First seen · 125 lines · 0 tokens per session scan A b3be64f18f86
PR Shepherd is an agent published in the GitHub repository boraoztunc/agent-system (11 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,984 tokens. 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.