functionality-scan

Reverse-engineers a codebase's features, entry points, workflows, and documentation gaps into structured findings.

3|1|Updated Aug 26, 2026
One-click install
npx skills add https://github.com/zeljkoobrenovic/sokrates-skills --skill functionality-scan-zeljkoobrenovic
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: functionality-scan
Source: https://github.com/zeljkoobrenovic/sokrates-skills/tree/main/skills/scanners/functionality-scan
Command: npx skills add https://github.com/zeljkoobrenovic/sokrates-skills --skill functionality-scan-zeljkoobrenovic

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams rarely have an accurate, code-grounded answer to "what does this software actually do" — the README describes aspirations, and no one has read every handler. This Skill produces a maintainer-grade functional description derived from the code itself, including hidden features and places where documentation and code disagree. ## Core Features & Use Cases - Feature and entry-point inventory: Enumerates CLI commands, routes, UI screens, hooks, and config keys from registration tables, then describes the 10-20 features that carry the product with evidence citations. - Workflow and data mapping: Reconstructs 3-6 end-to-end user workflows, inventories data stores and their lifecycles, and maps external integrations with their failure behavior. - Hidden functionality and doc-vs-code gaps: Finds feature flags, dormant commands, stubs, and every place the README or help text promises something the code does not do. - Use Case: During a handover or due diligence, run the scan against a repository with a Sokrates _sokrates analysis to get a validated functionality-scan.json report and an interactive explorer page describing what the software does. ## Quick Start Ask the AI to run the functionality scan on this repository and report what the software does, its main features, and where the documentation disagrees with the code.

Frequently Asked Questions about functionality-scan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find out what a codebase actually does?

Run the functionality scan, which reads registration tables like CLI subcommand enums, route tables, and config schemas to enumerate every entry point, then describes each feature's behavior, inputs, outputs, and failure modes with code citations.

How to check whether a README matches the actual code?

The scan reads documentation last and treats it as claims to verify, then produces hidden-and-gaps findings that cite both the documented promise and the code that contradicts it, in both directions.

Does the functionality scan require a Sokrates analysis?

It works best with a Sokrates _sokrates folder for component lists and config, and it depends on the sibling sokrates-scan-core skill for the findings format, validator, and renderer. Sokrates hotspot data can be skipped for this scanner.

What output does the functionality scan produce?

It writes _sokrates/reports/ai-insights/functionality-scan.json with grouped findings (purpose, features, entry-points, workflows, data, integrations, hidden-and-gaps), validates it, and renders an interactive HTML explorer page.

What are the limitations of a code-derived feature inventory?

Inferred workflow ordering is marked as likely rather than certain, and ambiguous gaps are marked possible since deliberate omissions can be misread as documentation errors. Depth is proportional to feature weight, so peripheral commands get lighter treatment.