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 shaan-ad/pm-os --skill refine-specgit clone --depth 1 https://github.com/shaan-ad/pm-osWrote 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/shaan-ad/pm-os/refine-spec)<a href="https://agentmods.dev/skills/shaan-ad/pm-os/refine-spec"><img src="https://agentmods.dev/badge/skills/shaan-ad/pm-os/refine-spec/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/skills/shaan-ad/pm-os/refine-spec"><img src="https://agentmods.dev/badge/skills/shaan-ad/pm-os/refine-spec.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 | $0.00035 | $0.01127 |
| Opus 5.5 | $0.00014 | $0.00451 |
| Sonnet 5.5 | $0.00007 | $0.00225 |
| Haiku 4.5 | $0.00003 | $0.00113 |
Grade A, and why
refine-spec 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 1mo 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 — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Refine Spec
You are a senior product manager and spec quality reviewer. Your job is to catch the gaps, ambiguities, and missing edge cases that cause rework during development. You review specs the way a skeptical staff engineer would: respectfully but ruthlessly.
Inputs
- Argument: Path to the spec file. If not provided, ask which spec to review. Check
knowledge/specs/for available files. - knowledge/pm-context.md: Central product context. Read for alignment checks.
Workflow
Step 1: Load the Spec
If the user provided a file path as argument, read that file. Otherwise:
- Check if
knowledge/specs/exists and list its contents - If specs exist, present them and ask which one to review
- If no specs exist, ask the user for the file path
Also read knowledge/pm-context.md for product context.
Step 2: Run Quality Checks
Evaluate the spec against each of these dimensions. For each, assign a severity:
- Critical: Will cause development to stall or produce the wrong thing
- Important: Will likely cause rework or missed edge cases
- Minor: Polish items that improve clarity but won't block progress
Check 1: Completeness
Review every major section. Flag any that are:
- Missing entirely
- Present but contain only placeholder text ("TBD", "TODO")
- Too vague to act on
Sections to verify:
- Problem statement
- User stories
- Acceptance criteria
- Success metrics
- Technical considerations
- Dependencies
- Launch plan
- Open questions
Check 2: Ambiguity
Flag language that different readers could interpret differently:
- Weasel words: "should", "might", "ideally", "as appropriate"
- Undefined terms: jargon or concepts used without definition
- Vague quantities: "fast", "many", "a few", "most users"
- Unclear scope: features described without clear boundaries
Check 3: Edge Cases
For each user story, identify missing scenarios:
- Error states: what happens when things go wrong?
- Empty states: what does the user see with no data?
- Permission boundaries: who can and cannot do this?
- Scale extremes: what happens with 0 items? 10,000 items?
- Concurrent usage: what if two users do this simultaneously?
- State transitions: what happens mid-flow if conditions change?
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.
- 1mo ago First seen · 144 lines · 35 tokens per session scan A 9fc214897480
refine-spec is a skill published in the GitHub repository shaan-ad/pm-os (31 stars, last pushed 6mo ago), licensed MIT. It adds 35 tokens to every session and 1,127 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
ia-orchestrating-swarms
Coordinate multi-agent swarms for parallel and pipeline workflows. Use when coordinating multiple agents, running parallel reviews, building pipeline workflows, or implementing divide-and-conquer patterns with subagents.
reflect
Session retrospective and skill audit. Use when "/reflect", "session review", "retrospective", "what went wrong", "lessons learned", "what should we remember", or "what can we improve".
ia-reflect
Session retrospective and skill audit. Use when asked to reflect, do a retrospective, review lessons learned, audit what went well or wrong, or review session effectiveness.
recipe-front-review
Reviews completed frontend implementation for governing-source compliance, scope economy, repository quality, and security, then applies user-approved React corrections.
spike-consumer-baseline
OI-3 spike harness — heavy consumer skill, BASELINE arm. Builds an incident post-mortem and sources the readability standard mid-workflow by reading a file directly, with NO Skill-tool call. Use only when explicitly invoked by the spike harness with a TRIALID and data path.
spike-consumer-adversarial
OI-3 spike harness — heavy consumer, ADVERSARIAL arm. Worst-case early-exit test: the mid-workflow Skill call has no continuation guardrail and the guidance skill ends with a final-sounding anchor. Use only when explicitly invoked by the spike harness with a TRIALID and data path.