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 skills add mshadmanrahman/pm-pilot --skill shepherd-learngit clone --depth 1 https://github.com/mshadmanrahman/pm-pilotWrote 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/mshadmanrahman/pm-pilot/shepherd-learn)<a href="https://agentmods.dev/skills/mshadmanrahman/pm-pilot/shepherd-learn"><img src="https://agentmods.dev/badge/skills/mshadmanrahman/pm-pilot/shepherd-learn.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00049 | $0.01057 |
| Opus 5 | $0.00024 | $0.00528 |
| Sonnet 5 | $0.00010 | $0.00211 |
| Haiku 4.5 | $0.00005 | $0.00106 |
Grade A, and why
shepherd-learn 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 yesterday.
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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/shepherd-learn — Capture Session Lessons
End-of-session command that captures what you learned into the institutional memory. This is what makes Bug Shepherd smarter over time.
Audience: Anyone who just finished a bug session. The lessons are written in plain language, not code.
Why This Matters
Every bug session teaches something. Without capturing it, you (or your team) will repeat the same mistakes, fall into the same traps, and waste time rediscovering the same patterns. The learning log is Bug Shepherd's institutional memory.
Workflow
1. Review the Session
Analyze what happened in this session:
- Which ticket was investigated? (
{TICKET-ID}) - What was the bug?
- What was the root cause?
- How many iterations did the fix take?
- Were there false starts or wrong assumptions?
- Did the reviewer catch anything?
- What was surprising or non-obvious?
2. Identify Lessons
Extract lessons in these categories:
Investigation Lessons:
- What did you look at first? Was that the right place?
- What assumption was wrong?
- What would you do differently next time?
Technical Lessons (even for non-developers):
- What component/feature area was the bug in?
- What made it hard to fix?
- Are there related areas likely to have similar bugs?
Process Lessons:
- Did the workflow help or hinder?
- Was context from the tracker sufficient?
- Did institutional memory surface useful past lessons?
Pattern Lessons:
- Is this a recurring bug pattern? (e.g., "long text overflows in translated locales")
- Does this bug suggest a systemic issue?
- Should a new safety rule be added to
triage.config.yaml?
3. Check for Repeated Mistakes
Read .claude/learning-log.md and check:
- Did this session repeat a mistake documented in a previous entry?
- If yes, flag it clearly:
"REPEATED MISTAKE: This is the same pattern as session {date} ({ticket})." "The lesson was: {previous lesson}" "Suggestion: Add a review check to catch this pattern."
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.
- yesterday Changed · +16 tokens per session 149739136df3
- 8d ago First seen · 146 lines · 33 tokens per session scan A 924a066a0f1d
shepherd-learn is a skill published in the GitHub repository mshadmanrahman/pm-pilot (20 stars, last pushed 2d ago), licensed MIT. It adds 49 tokens to every session and 1,057 once invoked, about $0.0002 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.
Other skills, from other repositories
triage-issue
Analyze a GitHub issue, verify claims against the codebase, and close invalid issues with a technical response.
hive.error-recovery
Follow a structured recovery decision tree when tool calls fail instead of blindly retrying or giving up.
manage-skills
A maintenance workflow for checking whether project verification skills still cover the code and rules that changed during a session.
systematic-debugging
Structured debugging methodology — use before proposing fixes for any error or failure. Covers: code bugs, build errors, deploy failures, config conflicts, dependency issues, infra problems. Also use when previous fix attempts failed or root cause is unclear.
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"…
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.