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 ianklassen/pm-skills --skill spec-from-conversationgit clone --depth 1 https://github.com/ianklassen/pm-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/skills/ianklassen/pm-skills/spec-from-conversation)<a href="https://agentmods.dev/skills/ianklassen/pm-skills/spec-from-conversation"><img src="https://agentmods.dev/badge/skills/ianklassen/pm-skills/spec-from-conversation/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/ianklassen/pm-skills/spec-from-conversation"><img src="https://agentmods.dev/badge/skills/ianklassen/pm-skills/spec-from-conversation.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 | $0.00108 | $0.00551 |
| Opus 5.5 | $0.00043 | $0.00220 |
| Sonnet 5.5 | $0.00022 | $0.00110 |
| Haiku 4.5 | $0.00011 | $0.00055 |
Grade A, and why
spec-from-conversation 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spec From Conversation
What this does
Takes messy input (a transcript, a thread, rambling notes) and produces a structured first-draft spec. This skill produces a draft to be argued with, not a final document. It should surface ambiguity the conversation left unresolved rather than silently resolving it for the reader.
Process
- Extract, don't invent. Pull the problem statement, goals, and constraints only from what was actually said. If the conversation didn't settle something (e.g., no one said what "done" looks like), write it into "Open Questions," don't guess a metric that sounds plausible.
- Separate stated goals from implied ones. If a stakeholder said "make it faster" but never gave a number, the spec should say "faster (no target given, needs follow-up)" rather than picking 200ms because it sounds reasonable.
- Draft non-goals explicitly, even if the conversation never mentioned them. A spec with no non-goals section is usually a spec that will grow scope silently later, so flag anything adjacent that was implicitly excluded (e.g., "web only, per the discussion; mobile wasn't mentioned either way, confirm").
- Flag conflicting statements rather than resolving them. If two people in the thread said different things, both go in with attribution, not a smoothed-over average.
Output structure
## Problem
[What's broken/missing, from the conversation, in plain terms]
## Goals
[What success looks like. Flag any goal stated without a measurable target]
## Non-Goals
[What's explicitly or implicitly out of scope]
## Success Metrics
[Only include metrics actually discussed; otherwise list under Open Questions]
## Open Questions
[Everything the conversation didn't resolve. This section should not be empty on a first draft]
What to watch for
Don't let this skill's output look more decided than the conversation actually was. A polished-looking doc with confident language creates false consensus, since stakeholders will assume agreement existed just because it's written down cleanly. When in doubt, under-resolve rather than over-resolve.
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 · 41 lines · 108 tokens per session scan A 0f924702cd96
spec-from-conversation is a skill published in the GitHub repository ianklassen/pm-skills (1 stars, last pushed 3mo ago), licensed MIT. It adds 108 tokens to every session and 551 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
feature-prioritisation
Aplica marcos de priorización (RICE, MoSCoW, Kano, ICE, Opportunity Scoring) para clasificar características e items del backlog. Úsalo cuando se te pida priorizar características, clasificar un backlog, decidir qué construir a continuación, o evaluar compensaciones entre ideas competidoras. Produce una lista de…
outcome-tracker
Record the testable predictions inside a decision, then score them against reality later — so frameworks earn trust from outcomes, not vibes. Use when committing to a prioritisation, forecast, or plan (to log what it predicts), when asked to review what actually happened, or to compute how well-calibrated past RICE…
tool-foundation-sprint-founding-hypothesis
Day 2 end capstone move of a Foundation Sprint. Compresses the sprint's full strategic frame into a single canonical sentence (the Founding Hypothesis) plus an assumption scorecard, why-we-believe, what-could-prove-us-wrong, and recommended next validation step. Use after Magic Lenses is signed. Strict canonical…
human-in-the-loop-design
Design the human approval surface for an agent system — which actions gate, how approvals batch without becoming rubber stamps, and what the audit trail must hold. Use when asked to add human oversight to an agent, design approval workflows for AI actions, decide what an agent may do autonomously, or fix approval…
agent-incident-postmortem
Run a blameless postmortem for an incident caused by an AI agent or LLM feature — hallucinated facts shipped to users, runaway tool use, prompt injection, cost blowouts, or wrong actions taken autonomously. Use when asked to write up an AI incident, analyse why an agent did something wrong, or produce corrective…
ai-assisted-performance-review
Evaluate performance fairly when output is AI-assisted — what still measures the human, what now measures the tooling, and how to run the review conversation. Use when reviewing someone whose work is heavily AI-assisted, when output volume stopped meaning anything, when calibrating a team with uneven AI adoption, or…