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 agentmods add skills/impactbrussels/ainativeos/apply-ai-native-modelsnpx skills add impactbrussels/AINativeOS --skill apply-ai-native-modelsgit 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/apply-ai-native-models)<a href="https://agentmods.dev/skills/impactbrussels/ainativeos/apply-ai-native-models"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/apply-ai-native-models.svg" alt="Measured on agentmods" 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 | $0.00149 | $0.01446 |
| Opus 5 | $0.00075 | $0.00723 |
| Sonnet 5 | $0.00030 | $0.00289 |
| Haiku 4.5 | $0.00015 | $0.00145 |
Grade A, and why
apply-ai-native-models 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 5d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Apply AI-Native Models
When you ask an AI to justify your idea, it will. That is the structural reason founders guess confidently and wrong: the easiest answer to reach is the one that flatters the plan. This skill makes you run a hard decision through a fixed set of the OS's mental models, in order, so the call is reasoned out and stress-tested rather than rationalised. It is the capstone over the whole build arc, from Frame to Defend.
The method
Run the decision through every model, in order, and refuse to cherry-pick. Full framework: references/mental-models.md. The models trace the build arc; when the memo points to a stage of work, route back through start-here.
Step 1: State the decision in one falsifiable sentence
Write the call as a claim that could be proven wrong, not a vibe. "Build our own retrieval layer" is a decision; "improve the product" is not.
Step 2: Score it against each model, in order
Run the full set. Each model is a one-line test and the trap it prevents. Do not stop at the first one that looks settled.
| Model | The test | The trap it prevents |
|---|---|---|
| Remove-the-AI test | Delete the AI: does the thing still break? | Building a feature and pricing it like a moat |
| Architecture-before-code | Is the system designed before any code? | The faster mess that collapses under growth |
| Adversarial-by-default | Have I made the case against, not just for? | Confirmation bias the model will happily feed |
| Human-on-the-loop | Is a human set above any irreversible step? | Automating a call that needs a signature |
| Store-the-proprietary, rent-the-replicable | Am I building what compounds, renting what does not? | Pouring effort into the commodity layer |
| A demo is not evidence | What did a real user, not a demo, prove? | Mistaking a working prototype for demand or safety |
| Proof before hype | What survives a web check or a real test? | Shipping a claim that cannot be defended |
| Trust as the currency | Does this earn or spend trust with the buyer? | Winning the demo and losing the market |
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.
- 5d ago First seen · 83 lines · 149 tokens per session scan A 6d18d7303b28
apply-ai-native-models is a skill published in the GitHub repository impactbrussels/AINativeOS (1 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 149 tokens to every session and 1,446 once invoked, about $0.0007 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.
agent-session-monitor
Real-time agent conversation monitoring - monitors Higress access logs, aggregates conversations by session, tracks token usage. Supports web interface for viewing complete conversation history and costs. Use when users ask about current session token consumption, conversation history, or cost statistics.
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.
nw-ddd-eventsourcing
Event Sourcing and CQRS as DDD implementation patterns — when to use, aggregate event streams, projections, snapshots, sagas, upcasting, conflict resolution.
nw-leanux-methodology
LeanUX backlog management methodology - user story template, story sizing, story states, task types, Definition of Ready/Done, anti-pattern detection and remediation.