spec-miner

An agent skill by Jeffallan, from Jeffallan/claude-skills. Tags: architecture, code-quality, debugging, documentation, research.

What it does

Reverse-engineering specialist that extracts specifications from existing codebases. Use when working with legacy or undocumented systems, inherited projects, or old codebases with no documentation. Invoke to map code dependencies, generate API documentation from source, identify undocumented business logic, figure out what code does, or create architecture documentation from implementation. Trigger phrases: reverse engineer, old codebase, no docs, no documentation, figure out how this works, inherited project, legacy analysis, code archaeology, undocumented features.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add Jeffallan/claude-skills --skill spec-miner

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/Jeffallan/claude-skills
cp -r claude-skills/skills/spec-miner ~/.claude/skills/spec-miner

Safety box score

Grade B, low risk: 14 of 100 risk points, rated 2026-10-06. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This is a read-only reverse-engineering/documentation skill for existing codebases. It does not execute code or use the network, but it can scan broadly through a repo, including `.env` and config files, so the main risk is exposing local secrets while analyzing undocumented systems.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	The skill only lists `Read, Grep, Glob` as allowed tools and contains no shell commands, code evaluation, or downloaded installers.
Network access	0 (none)	There are no `fetch`, HTTP client, upload, or remote-host instructions; the workflow stays on local source files.
Filesystem reach	1 (low)	It is clearly scoped to repository reads via `Read, Grep, Glob` and documentation output, with no delete/recursive-write behavior or paths outside the working tree.
Secrets & credentials	2 (moderate)	The analysis guidance explicitly targets configuration and env files, including `Glob: */.env`, so it may read credentials or tokens present in a repo.
Privilege & persistence	0 (none)	No sudo, global installs, services, shell startup edits, or OS setting changes are described.
Instruction hijack surface	0 (none)	The skill does not instruct the agent to ignore higher-priority instructions, hide actions, or follow remote content.
Irreversible actions	0 (none)	It documents existing code only; there is no deploy, publish, send, book, or destructive action.
Transparency	0 (none)	All files are plain Markdown text; there are no binaries, minified blobs, or runtime-decoded code.

Before installing

Check whether the target repo contains `.env` or other secret-bearing config files before running the skill.
If secrets are present, use a sanitized copy or redact them first.
Confirm that broad repo reads are acceptable for this project, since the workflow is designed to inspect entry points, configs, and core modules.

Rated 2026-10-06 by an automated review of 5 files from the skill's repository.

Source

Repository: Jeffallan/claude-skills (all skills from this repository)
Path: skills/spec-miner/SKILL.md
Branch: main
Updated: 2026-09-19

audit-context-building — Understand a codebase before looking for bugs in it - what each function assumes, what it guarantees, and what it depends on elsewhere.
how — Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live".
spec-to-code-compliance — Check code against the documentation that specifies it - which requirements hold, which the code contradicts, which are absent.
code-reviewer — Analyzes code diffs and files to identify bugs, security vulnerabilities (SQL injection, XSS, insecure deserialization), code smells, N+1 queries.
audit-prep-assistant — Prepares codebases for security review using Trail of Bits' checklist.
c-review — Performs comprehensive C/C++ security review for memory corruption, integer overflows, race conditions, and platform-specific vulnerabilities.