data-consumer-discovery — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited data-consumer-discovery (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Ask these in order. Each builds on the prior answer:
NEVER ask "what data do you want?" or "what would you like us to build?" These questions produce wish lists, not validated needs. The Mom Test applies to data teams: talk about their life, not your product.
The strongest discovery signal is existing workarounds. If an analyst built a 47-tab Excel workbook, that's a validated need with proven demand.
For every workaround you find, document:
Workarounds with daily frequency and downstream dependents are the highest-signal discovery findings. Build for these first.
Data consumers fall into four segments. Each needs different product shapes:
| Segment | Need | Product Shape |
|---|---|---|
| Explorers | Flexible query access, iterate fast | Self-serve query tools, semantic layer |
| Reporters | Scheduled, consistent outputs | Automated reports, dashboards with alerts |
| Decision-makers | One number with context | Executive summaries, scorecards, red/yellow/green |
| Builders | Reliable APIs and tables | Documented APIs, data contracts, SLAs |
ALWAYS identify which segment you're building for before writing requirements. A product that tries to serve all four serves none.
Ask: "What would make you NOT trust this output?"
Common trust barriers in data products:
Map each barrier to a specific data-quality-assessment dimension. Trust barriers are quality problems experienced from the consumer's perspective.
Rank evidence before acting on it:
NEVER build on Emerging signals. Moderate signals justify a 1-day experiment (see data-product-validation). Strong signals justify a shaped pitch.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.