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 impactbrussels/AINativeOS --skill measure-ai-native-pmfgit clone --depth 1 https://github.com/impactbrussels/AINativeOSWrote 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/impactbrussels/ainativeos/measure-ai-native-pmf)<a href="https://agentmods.dev/skills/impactbrussels/ainativeos/measure-ai-native-pmf"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/measure-ai-native-pmf/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/impactbrussels/ainativeos/measure-ai-native-pmf"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/measure-ai-native-pmf.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 | $0.00156 | $0.01233 |
| Opus 5 | $0.00078 | $0.00616 |
| Sonnet 5 | $0.00031 | $0.00247 |
| Haiku 4.5 | $0.00016 | $0.00123 |
Grade A, and why
measure-ai-native-pmf 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 9d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Measure AI-Native PMF
Vanity metrics are dead. Total users and total traffic climb whether or not a single person came back, and AI made that lie cheaper to tell: a model writes the onboarding, a launch post lands, ten thousand people try the thing in a week, and the graph looks like fit. It is curiosity with good distribution. The first purchase measures curiosity. The fifth measures belief. This skill reads the gap.
The method
Four lenses, drawn from handbook chapter 12. Run them in order; behaviour and money outrank anything the founder states they believe. Full method, the cohort-signal table, a worked PMF read, anti-patterns, and a copyable template: references/pmf-method.md.
-
Run the 40% test. Survey active users, not the sign-up list, with the Sean Ellis question: how would you feel if you could no longer use the product? More than 40% answering "very disappointed" is the bar. It measures dependence, not enthusiasm; a five-star rating from someone who never returns is not fit.
-
Run the Remove-the-AI test on the fit, not the stack. Ask what repeat buyers are paying to keep. If the product still does its job once the model is gone, you measured demand for a feature anyone can rent. The part they are loyal to has to be the part that breaks without your data.
-
Measure Share of Model. Run 20 to 50 real buyer queries across ChatGPT, Claude, and Perplexity. Record how often each answer cites you, a competitor, or neither. That ratio is the new top of the funnel: visibility in the channel where considered buying now starts. Watch the trend, not the snapshot.
-
Check the second bite. Pull the retention curve. Did the January cohort survive to May, did the line flatten into a floor, did the reorder happen without a coupon? A spike in trial with flat repeat is a warning dressed as a win. Trial is the cost of finding out; habit is the business.
Output
- A four-lens PMF read: the 40% score, the Remove-the-AI verdict (native or borrowed retention), the Share-of-Model ratio, and the second-bite curve, each stated plainly.
- The one competitor beating you on Share of Model, with a one-line hypothesis why.
- Next: run
capture-learningon the result so the OS records what the number actually proved.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 67 lines · 0 tokens per session scan A 00b28ef462f1
measure-ai-native-pmf is a skill published in the GitHub repository impactbrussels/AINativeOS (1 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 156 tokens to every session and 1,233 once invoked, about $0.0008 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 2026-08-31.
Other skills, from other repositories
remember
Review reusable project knowledge and decide what belongs in project memory, notepad, or durable docs.
docx-comment-reply
Reply to comments (批注) in Word .docx/.doc files: extract comment context, draft replies, write threaded replies back, and validate OOXML.
explaining-machine-learning-models
Explain trained machine learning models through feature attribution, local explanations, and behavior summaries. Use as an explicit/manual helper once a model already exists, not for training ownership, leakage auditing, or general ML strategy selection.
config-validator
Validate AIWG configuration files and project setup for correctness and completeness.
aiwg-regenerate-copilot
Regenerate copilot-instructions.md for GitHub Copilot with vendor-specific content only.
checkpoint
Create, list, or recover mid-workflow checkpoints so interrupted work resumes from a known-good position.