- Home
- Skills
- Research & Analysis
- discovery audit
Works with the AI Agents you already use
discovery audit
Audit application code against product claims to verify implemented features and identify security or maturity gaps.
$29
discovery audit
Example session with this skill installed
Run a discovery audit on the current repo against the claims in brief.md. Use the document-signing profile and output the P0 findings.
- Read your context and instructions
- Compiled the discovery audit
I have completed the discovery audit for the application against docs/remediation/work/discovery/brief.md. I processed two main areas: Core Logic and Document Flow (Signature Concepts).
The audit identified 14 implemented capabilities, 4 partial implementations, and 3 high-priority findings.
Discovery Audit Summary
| ID | Area | Capability | State | Evidence Path |
|---|---|---|---|---|
| CAP-101 | Core | JWT Authentication | [x] | src/auth/jwt.py -> src/middleware/auth.mw |
| CAP-102 | Core | Role Based Access | [~] | Missing test coverage for 'Editor' role |
| CAP-201 | Sign | PAdES Signature | [x] | src/sign/pades.py -> tests/test_pades.py |
| CAP-202 | Sign | LTV Verification | [ ] | Absent: No logic found in src/sign/ |
Critical Findings (P0/P1)
| Finding ID | Priority | Type | Description |
|---|---|---|---|
| F-DSC-101 | P0 | Fact | Hardcoded secret in config/production.yaml |
| F-DSC-201 | P1 | Inference | Signature timestamping lacks external NTP sync |
Product Maturity: Level 2 (Functional)
The core signature engine is implemented, but the product overstates its LTV (Long Term Validation) capabilities in the brief.
Next steps
- Review the full report at
docs/remediation/discovery-audit.md. - Run
evolution-auditto assess technical debt in the implemented areas. - Invoke
action-planto address the P0 hardcoded secret finding.
Connects securely to your tools. The creator never sees your data.
About this skill
Technical due diligence is often stalled by the gap between what a product brief claims and what the codebase actually delivers. This skill systematically audits applications to verify every implemented feature against end-to-end evidence in code, configuration, and tests.
What it does
- Verify claims by matching product brief requirements against live code, identifying unimplemented or scaffolded features.
- Generate technical inventory including capability rows (CAP-nnn) and defect findings (F-DSC-nnn) with maturity ratings.
- Detect security gaps by flagging P0–P3 security and reliability findings based on a formal inspection checklist.
- Assess product maturity across four standardized levels to determine if a product is ready for white-labeling, API exposure, or sale.
How it works
- Initialize brief by saving the product claims to
brief.mdand selecting audit areas from standardized profiles or custom checklists. - Execute audit using a
product-auditorrole to inspect code, using the provided Python scripts to check for evidence of entry points and persistence. - Merge findings into a master
discovery-audit.mdreport using theaudit.pyutility to aggregate area-specific data and maturity levels. - Validate report against a strict quality gate that ensures every implemented feature cites a specific end-to-end path rather than just a lone file.
Frameworks & tools
This skill requires Python 3 for its core scripts (audit.py, decisions.py, page.py). It is designed to work with local repositories and supports subagent orchestration if available in the host environment.
Why this beats prompting it yourself
Standard prompts often hallucinate functionality by seeing a class name and assuming the feature works. This skill enforces a "no evidence, no credit" rule, requiring a trace from entry point to persistence before marking a capability as implemented.
Use cases
- Conduct technical due diligence before an acquisition or investment.
- Audit a product baseline before starting a major white-label or API project.
- Verify "ready-to-sell" status for new features or modules.
- Identify documentation that overstates the product's actual capabilities.
Known limitations
Requires Python 3 for bundled scripts. The audit is inspection-only and will not execute code, modify databases, or reproduce secrets. Briefs are required to establish the baseline for measurement.
Requires Python 3. Inspection-only: cannot modify files or run dynamic analysis. Relies on a provided product brief to measure implementation gaps.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 12 days ago
Requires