skills.rest
tense-i
tense-iCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

reverse-spec-from-repo

Reverse-engineer a prompt-spec suite from a codebase into spec/ with evidence.

1|Updated Mar 8, 2026
Search Tags:#handoff#prompts#workflows#repo-analysis#reverse-engineer#evidence-led#prompt-spec
One-click install
npx skills add https://github.com/tense-i/stock-market-simulator --skill reverse-spec-from-repo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: reverse-spec-from-repo
Source: https://github.com/tense-i/stock-market-simulator/tree/main/skills/reverse-spec-from-repo
Command: npx skills add https://github.com/tense-i/stock-market-simulator --skill reverse-spec-from-repo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reverse-engineers a usable prompt-spec suite from an existing codebase, turning scattered code, tests, and documentation into a traceable handoff artifact that clearly separates observed behavior, inferred intent, and unresolved ambiguity.

Core Features & Use Cases

  • Produces a complete spec package under spec/ including README.md, context-manifest.md, evidence.md, and structured prompts under prompts/, modules/, and workflows/ grounded in repo evidence.
  • Distinguishes observed signals from inferred intents and open questions to support auditability and easy handoffs to coding agents.
  • Supports scoped reverse-spec generation (whole-repo, subsystem, or feature-path) to fit project size and risk.

Quick Start

Create a reverse-spec run for the target repository and generate the spec package under spec/.

Frequently Asked Questions about reverse-spec-from-repo

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

FAQPage Schema
How does evidence-led repo analysis work for generating handoff artifacts?▼

Evidence-led repo analysis extracts observed signals from code and tests, then maps them to inferred intents and open questions, producing an evidence ledger and structured prompts tied directly to repository evidence.

Can I generate a prompt-spec for a specific subsystem instead of a whole monorepo?▼

Yes, you can scope reverse-spec generation to a whole-repo, subsystem, or feature-path, allowing the prompt-spec generation to fit project size and risk for both small utilities and large monorepos.

What files are included in a reverse-engineered spec package for coding agents?▼

A reverse-engineered spec package includes README.md, context-manifest.md, evidence.md, and structured prompts organized under prompts/, modules/, and workflows/ directories, all grounded in repository evidence.

What is the best way to document unresolved ambiguity when reverse-specifying a repository?▼

The best way to document ambiguity is by maintaining an explicit separation of observed signals, inferred intents, and open questions within an evidence ledger, ensuring auditability and clear handoffs to coding agents.

When should I not use reverse-spec generation for a codebase handoff?▼

You should avoid reverse-spec generation when a repository lacks sufficient code, tests, or documentation to extract meaningful evidence, as the process requires observable signals to distinguish inferred intents and open questions.

Related Skills

Explore More Software Engineering Skills→
shanyan-wcxshanyan-wcx
96

trellis-spec-bootstarp

Analyze repository code patterns to generate .trellis/spec files and index.

Community
Intermediate
yz5eyz5e
15

forge

Generate XML-tagged prompts for coding agents via structured interviews.

Community
Advanced
tonykooptonykoop

reverse-engineer

Document unknown artifacts by separating observed facts from inferences with confidence levels.

Community
Advanced
vikisingh23vikisingh23
4

documentation

Generate API references, architecture diagrams, and onboarding guides from code and OpenAPI specs.

Community
Intermediate