Detect contradictions between documentation, specifications, and code.
npx skills add https://github.com/solatis/claude-config --skill incoherence
Please help me install this Agent Skill. Skill: incoherence Source: https://github.com/solatis/claude-config/tree/main/skills/incoherence Command: npx skills add https://github.com/solatis/claude-config --skill incoherence
Incoherence detection and reconciliation to resolve contradictions between docs, specs, and implementation, with a structured workflow to apply user resolutions.
Start the detection workflow: python3 scripts/incoherence.py --step-number 1 --total-steps 22 --thoughts "Analyzing project X"
High-intent search queries and answers about installing and using this skill.
Documentation-code incoherence detection identifies contradictions, ambiguities, and gaps across docs, specifications, and implementation through a structured 22-step workflow that maps inconsistencies across multiple dimensions and produces a reconciliation report.
Use a coordinated detection and reconciliation workflow that explores inconsistencies in parallel, catalogs problematic dimensions, and applies user-provided resolutions iteratively to ensure updated artifacts remain coherent across all project layers.
Yes, the Python-based workflow accepts a user-specified report filename and coordinates parent and sub-agent phases to automatically detect mismatches, generate structured issue reports, and apply reconciliation—supporting idempotent re-runs as specs and code evolve.
The workflow applies to software projects containing documentation files, formal specifications, and source code; it accepts a report filename parameter and outputs a structured issue-resolution report with updated, internally consistent artifacts.
The 22-step process maps potential inconsistencies across a dimension catalog, explores them in parallel, and synthesizes resolutions that coordinate parent and sub-agent phases to recheck outcomes and ensure all artifacts stay synchronized.
The workflow requires explicit user resolutions for detected incoherence; it works best on projects with distinct documentation, specifications, and implementation layers, and relies on user-provided report filenames to structure output consistently.
RyderFreeman4LogosAnalyze Rust crates to generate facts.toml, README.md, review_report.md, and blog.md.
patrickrmooreDocument and enforce project principles governing SDD workflow decisions.
paulojalowyjRun linters, type checkers, and security audits for Node.js and Python projects.
Rubiss-ProjectsAutomates Git commits triggered by Spec Kit command events.
