More screenshots
Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
GitHub Copilot
Gemini CLI+20 moreClaude CodeCursorCodex CLIGitHub CopilotGemini CLI
OpenClaw
Manus
Grok+17 agentsMarketplace Research Evidence Auditor
Audit AI marketplace research for unsupported claims, duplicate evidence, coverage gaps, and the next evidence to obtain.
$5
Marketplace Research Evidence Auditor
Example session with this skill installed
Audit these marketplace research reports against my requirements. Show unsupported conclusions, compare the candidates, and tell me what existing evidence to provide or what to search next.
- Read your context and instructions
- Compiled the marketplace research evidence
- Generated the document
Evidence audit result: VERIFY
The submitted reports contain useful market evidence, but the current shortlist is not ready for a clean decision. The calculations below reconcile; several conclusions still need source support or clarification.
| Research claim | Audit finding | Required next action |
|---|---|---|
| Moving planner: STRONG. | Listing counts and several calculations are supported, but claims about a clear product boundary and missing competitor features are not tied to supplied sources. | Provide the competitor feature evidence before committing to production. |
| Bill tracker: median price below $8 means pricing is unhealthy. | Your requirements contain no minimum price. The report introduced an extra decision rule, so this remains an unresolved inference. | Clarify whether a minimum price belongs in the decision criteria. |
| Two pages were checked for each candidate. | The audit kept 5 new query calls and 3 reused-page records separate. The reused first pages still need their original source material. | Provide those existing first-page reports; do not run another search yet. |
Calculation checks
- Four reported price medians reconciled with the source-linked values.
- Four reported single-shop revenue shares also reconciled.
- Correct arithmetic did not clear the open evidence and coverage findings.
- All four candidates therefore remain
VERIFY. VERIFY means more evidence or clarification is required; it does not mean the markets have no demand.
Next evidence required
- Provide the existing first-page material referenced by the three reuse records.
- Add sources for the claimed product boundaries and missing competitor features.
- Confirm whether a minimum acceptable price is actually one of your rules.
New searches: NONE
The current gaps concern existing material and an undefined criterion. No additional marketplace search was executed.
marketplace-research-evidence-auditor.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
Make your AI prove the opportunity
AI market research can look convincing while relying on duplicated shops, incomplete coverage, unsupported conclusions, or criteria you never approved. This Skill audits the reports and requirements you provide before you commit to building a product.
What it checks
- Separates listings from independent shops and keeps conflicting or excluded evidence visible.
- Checks query, page, and sample coverage without counting reused material as new research.
- Reconciles supported counts, medians, percentages, and source-linked corrections.
- Flags buyer-job mismatches, undefined fields, unsupported inferences, and unresolved evidence.
- Returns KEEP, VERIFY, or REJECT under your supplied rules and explains what evidence to provide next.
What you receive
The ZIP contains the Skill, two local Python tools, and reference instructions. Every completed audit produces a readable summary, evidence ledger, detailed findings, a bounded next-query file, and an input/output receipt.
How to use it
- Install the Skill in an AI assistant that can access local files and run Python.
- Provide TXT or Markdown research reports, pasted text, and your decision requirements.
- Ask the Skill to prepare and audit the evidence. No JSON authoring is required.
Requirements and limits
Requires local file access and Python. Validation used Windows with Python 3.14.3; every host and operating system has not been verified. A fileless web chat alone cannot run the bundled tools. The tools do not search marketplaces, use the network, overwrite prior audits, or guarantee demand or sales. Semantic review depends on the host AI and may miss or misinterpret evidence. KEEP means the submitted evidence satisfies the supplied rules; it is not a sales prediction.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 23 days ago
- Passed all security checks, Safe to install