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 customer-discovery-that-doesnt-liegit 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/customer-discovery-that-doesnt-lie)<a href="https://agentmods.dev/skills/impactbrussels/ainativeos/customer-discovery-that-doesnt-lie"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/customer-discovery-that-doesnt-lie/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/customer-discovery-that-doesnt-lie"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/customer-discovery-that-doesnt-lie.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 | $0.00125 | $0.01310 |
| Opus 5 | $0.00063 | $0.00655 |
| Sonnet 5 | $0.00025 | $0.00262 |
| Haiku 4.5 | $0.00013 | $0.00131 |
Grade A, and why
customer-discovery-that-doesnt-lie 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 10d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Customer Discovery That Doesn't Lie
Talk is cheap, and AI made it cheaper. The people you interview are polite, and the model you hand the transcripts to is agreeable, so two lies stack into one dangerous object: a discovery process that feels rigorous and confirms exactly what you hoped. The friction that used to protect you, the slow read of every transcript by hand, is gone. Design the disconfirmation back in on purpose, or your research will only ever grant you the permission you came for.
The method
Discovery built to disconfirm, in the Mom-Test posture. Full framework: Handbook
Chapter 04. Deep method, tables, and a worked example:
references/discovery-method.md.
Step 1: Write a script that asks about the past, not the hypothetical
A good question is one the customer cannot answer with a compliment. Ask what they did, paid for, and worked around last month, never what they would do with your idea.
| Bad question (invites a lie) | Good question (extracts a fact) |
|---|---|
| "Would you use a tool that did X?" | "Walk me through the last time you hit [PROBLEM]. What did you do?" |
| "Do you think this is a good idea?" | "What have you tried to fix this? What did you pay for it?" |
| "How much would you pay?" | "What does the current workaround cost you in time or money?" |
Step 2: Interview to advance, not to be liked
A compliment is not data. Seek a commitment that costs the customer something: time, a referral, access, money, a follow-up booked. No advance means no real interest, however warm the call felt.
Step 3: Extract facts from raw transcripts, never a mood summary
Feed the model the raw transcripts and forbid the average. Make it pull every concrete claim and tag who said it, then score each pain on four axes: severity, frequency, current workaround, and willingness-to-pay signal. The signal lives in the edge case, not the centre of mass.
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.
- 10d ago First seen · 91 lines · 125 tokens per session scan A 2d494773fa69
customer-discovery-that-doesnt-lie is a skill published in the GitHub repository impactbrussels/AINativeOS (1 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 125 tokens to every session and 1,310 once invoked, about $0.0006 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.