skills.rest
gleeb
gleebCOMMUNITY
Explore all skills by this author →
GitHub

semantic-review

Validate local-model outputs and code quality against story scope via integrity, semantic, and terminology checks.

Updated Aug 23, 2026
Search Tags:#code quality#acceptance criteria#evidence verification#semantic review#terminology alignment#git diff scoping
One-click install
npx skills add https://github.com/gleeb/ai-registry --skill semantic-review
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: semantic-review
Source: https://github.com/gleeb/ai-registry/tree/main/systems/sdlc/skills/semantic-review
Command: npx skills add https://github.com/gleeb/ai-registry --skill semantic-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Semantic review verifies that local-model outputs and the implemented code truly match the planned scope and acceptance criteria, reducing drift, confabulation, and unreliable “claimed verification” after changes are integrated.

Core Features & Use Cases

  • 3-check validation workflow that defaults checks to NEEDS WORK until proven PASS.
  • Agent report integrity to detect contradictions within and across reviewer/QA reports and ensure all agents operated on the same scope.
  • Senior-developer code quality review that drills into every changed file, traces logic paths, and re-runs evidence verification commands from the reports.
  • Terminology and contract alignment to catch naming drift between contracts/architecture/stories and the actual implementation.
  • Structured guidance packages that produce actionable correction steps and knowledge-gap fetch instructions for re-dispatch when issues are found.

Quick Start

Use the semantic-review skill when the sdlc-engineering-semantic-reviewer needs to validate local model outputs and review the implemented code quality for a completed story before acceptance validation.

Frequently Asked Questions about semantic-review

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

FAQPage Schema
How do I verify code quality against acceptance criteria after integrating changes?▼

Code quality verification checks implemented code against story scope and acceptance criteria using a 3-check workflow. It runs integrity, semantic, and terminology alignment checks with git diff scoping, defaulting every check to NEEDS WORK until evidence proves PASS.

What is semantic review in the software development lifecycle?▼

Semantic review validates that local-model outputs and implemented code match planned scope, reducing drift and confabulation. It detects contradictions within and across reviewer reports, ensuring all agents operated on the same scope before acceptance validation.

How do I check for terminology and contract drift in implemented code?▼

Terminology alignment checks catch naming drift between contracts, architecture, stories, and the actual implementation. By drilling into every changed file and tracing logic paths, the review identifies mismatches between documented contracts and code reality.

Does code review work with git diff scoping for staging documents?▼

Yes, semantic review uses git diff scope alongside a staging document and full file drill-down. It applies during SDLC execution-phase dispatches after story integration and QA, re-executing evidence verification commands from reports.

What is the best way to generate actionable correction steps after finding code discrepancies?▼

Structured guidance packages produce actionable correction steps and knowledge-gap fetch instructions for re-dispatch when issues are found. These packages include proactive PASS observations and documentation guidance driven by identified knowledge gaps.

When should I not use automated semantic review for code validation?▼

Automated semantic review is designed for SDLC execution-phase dispatches after story integration and QA. It targets local-model outputs and implemented code validation, making it unsuitable for pre-integration planning or non-story-based development contexts.

Related Skills

Explore More Software Engineering Skills→
CarlosQ96CarlosQ96

qa-plan

Generate structured QA plans from GDDs and story files.

Community
Advanced
danilonovaisvdanilonovaisv
1

subagent-driven-development

Dispatch isolated subagents per task and run spec-compliance and code-quality reviews.

Community
Advanced
ActiveInferenceInstituteActiveInferenceInstitute
30

gnn-validation

Validate GNN models across semantic, structural, and mathematical checks.

Community
Advanced
olimpus-softolimpus-soft

review-py

Review Python code diffs using templates, checklists, and severity criteria.

Official
Advanced