Docs analysis

Skill ellmos-ai/skills/skills/dev/docs-analysis

Portable SKILL.md library for Claude Code, Codex-compatible agents, BACH, and local-first LLM workflows

Install

npx -y skills add ellmos-ai/skills --skill docs-analysis

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Document requirements analysis: Analyzes concept and requirements documents in the docs/ folder, checks requirements against current code, and creates a consolidated difference report.

SKILL.md

3.3 KB, as published. Nobody here has run it

Deutsch — Offizielle Deutsch-Version / Documento Oficial en Deutsch.

Document Requirements Analysis (Deutsch)

Analyzes all concept and requirements documents, checks their requirements against the current code, and creates a consolidated difference report.

Übersicht & Zweck

Analyzes all concept and requirements documents in the ../docs/ folder, checks their requirements against the current code, and creates a consolidated difference report.

Naming Convention

Prefix and Suffix

All analyzed documents receive:

Prefix: conN_ where N = analysis version (1, 2, 3, ...)
Suffix: _XX where XX = fulfillment percentage (rounded to nearest 10)

Archiving Threshold

>= 75% fulfilled: Document is moved to ../docs/_archive/
< 75% fulfilled: Document stays in ../docs/ with prefix/suffix
Threshold configurable (default: 75)

Process

Phase 1: Collect Documents

List all *.md and *.txt files in ../docs/ (root)
Filter out README.txt

Phase 2: Extract Requirements

For each document:

Read content
Identify requirements (checklists, tables, MISSING/TODO markers)
Categorize: Structure, Code, API, DB Schema, CLI, Feature

Phase 3: Code Verification

For each requirement:

Determine verification method (Glob, Grep, Read)
Execute verification
Mark as: FULFILLED, PARTIAL, MISSING

Phase 4: Assessment

Count fulfilled vs. open requirements
Calculate fulfillment percentage (%)
Decide: archive (>= 75%) or keep (< 75%)

Phase 5: Generate Output

Create REQUIREMENTS_ANALYSIS.md (summary)
Create consense_diff.md (only open requirements, by priority)

Phase 6: Versioning

Scan for highest conN_ prefix
New version = highest + 1

Phase 7: Rename and Move

Apply new prefix/suffix to documents
Archive or keep

Output

File	Description
`conN_REQUIREMENTS_ANALYSIS.md`	Complete analysis (version N)
`consense_diff_N.md`	Consolidated open requirements
`_archive/conN__XX.`	Archived (>=75%) documents

Priority Classification

Priority	Criteria
P1	Core functionality missing, system unusable
P2	Important feature missing, workaround possible
P3	Nice-to-have, improves UX
P4	Cosmetic, documentation, code quality

Änderungsprotokoll

1.0.0 (2026-03-15)

Ported from BACH v3.8.0

Ported from BACH v3.8.0 | Standalone Version