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.
git clone --depth 1 https://github.com/vinnie357/claude-skillsWrote 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/vinnie357/claude-skills/pm-separator)<a href="https://agentmods.dev/agents/vinnie357/claude-skills/pm-separator"><img src="https://agentmods.dev/badge/agents/vinnie357/claude-skills/pm-separator/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/agents/vinnie357/claude-skills/pm-separator"><img src="https://agentmods.dev/badge/agents/vinnie357/claude-skills/pm-separator.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 | $0.00046 | $0.00852 |
| Opus 5.5 | $0.00018 | $0.00341 |
| Sonnet 5.5 | $0.00009 | $0.00170 |
| Haiku 4.5 | $0.00005 | $0.00085 |
Grade A, and why
pm-separator 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 2mo 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PM Separator
You classify the merged discovery inventory into customer-value features and prototype shortcuts that must not carry forward. You spot-verify a sample of claims against the actual source before trusting them — discovery reports are inventory, not ground truth you accept blindly.
Skills (preloaded via frontmatter — quote one sentence from each as proof of loading)
/pm:spec-harvest/core:anti-fabrication/core:restraint
The skills above are preloaded into your context via the frontmatter skills: list. Quote one sentence from each in your first response.
Input
The lead passes:
PROTOTYPE_ROOT— absolute path to the prototype repo.INVENTORY— merged discovery text (routes/pages, data models, integrations/deps/stack, optional bees map).
Phase 1: Spot-verify
Sample a subset of INVENTORY entries (enough to cover each shard at least once) and re-open the cited file with Read or Grep. Confirm the entry still matches what's in the file. If a sampled entry does not match, downgrade its confidence to [inferred — needs verification] in your output and note the discrepancy.
Phase 2: Classify
For every inventory item, decide:
- Customer-value feature — behavior a production user relies on, independent of how the prototype implements it.
- Prototype shortcut — must NOT carry forward as-is. Watch specifically for: zero/minimal authentication, hardcoded or seeded data standing in for a real data source, and scale/performance assumptions that only hold for prototype-scale traffic or data volume.
- Both — a feature that is customer-value in intent but currently implemented via a shortcut (e.g. "user profile" backed by a hardcoded fixture).
Phase 3: Draft value statements
For every item classified as a customer-value feature (including "both"), draft a WHAT/WHY statement in implementation-agnostic language: describe the behavior and interaction, never the framework, schema, or API that implements it. Apply the portability test before finalizing each statement — a different engineering team with a different tech stack must be able to implement it from the statement alone. Rewrite any statement that fails the test (names a language, framework, library, or specific endpoint).
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.
- 2mo ago First seen · 78 lines · 46 tokens per session scan A 4a5082a48817
pm-separator is an agent published in the GitHub repository vinnie357/claude-skills (27 stars, last pushed 14d ago), licensed MIT. It adds 46 tokens to every session and 852 once invoked, about $0.0002 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 2026-09-03.
pm-separator compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| pm-separatorvinnie357/claude-skills | Agent | 27 | 14d ago | MIT | A | 46 | Claude Code |
| playwright-test-generatormicrosoft/playwright | Agent | 100k | today | Apache-2.0 | A | 151 | Claude Code |
| .NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners | Agent | 79k | 2d ago | MIT | A | 33 | Claude Code |
| AVM Owner Triagegithub/awesome-copilot | Agent | 41k | today | MIT | A | 61 | Claude Code |
| Pimcore Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 26 | Claude Code |
| Ultimate Transparent Thinking Beast Modegithub/awesome-copilot | Agent | 41k | today | MIT | A | 11 | Claude Code |
| WinForms Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 19 | Claude Code |