Skill: module-bound-check

SkillDev tools

Run all 5 decomposition validators (req-has-module, code-annotation-maps-to-module, visibility-rule-enforcement, anaemic-context-detection, granularity-match) on the project, reporting violations.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Skill: module-bound-check skill

What this skill tells your AI

The instructions your AI receives, as published by stevegjones/ai-first-sdlc-practices in skills/module-bound-check/SKILL.md and read by ahel’s review.

Use this skill to verify the project's decomposition is sound. Aggregates results from all 5 module-bound validators and reports a single pass/fail with per-validator detail. Run this before pushing a substantial change to a regulated-context project.

Steps

  1. Load the decomposition. Read programs.yaml (or use default_decomposition if absent).

  2. Build the spec_module_lookup. Walk all docs/specs/**/*.md artefacts; map each declared ID → module (from frontmatter or positional ID prefix).

  3. Build the import-edge graph. Walk source files; for each import, determine the from-module and to-module by mapping the file path to declared module paths.

  4. Run each validator.

    • req_has_module_assignment
    • code_annotation_maps_to_module
    • visibility_rule_enforcement (mode from team-config: strict for regulated, advisory otherwise)
    • anaemic_context_detection
    • granularity_match
  5. Aggregate results. Print a table:

    Validator                          Status   Errors  Warnings
    req_has_module_assignment          PASS     0       0
    code_annotation_maps_to_module     FAIL     2       0
    visibility_rule_enforcement        PASS     0       1
    anaemic_context_detection          PASS     0       0
    granularity_match                  PASS     0       3
    
  6. Report each error and warning with file path and line number.

Done criteria

  • All 5 validators ran.
  • Result table printed.
  • Exit code: 0 if all pass; 1 if any error; 0 if only warnings (warnings don't block).

Signals

GitHub stars
41
Forks
6
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
module-bound-check
Source
github.com/stevegjones/ai-first-sdlc-practices