Skill Claude CodeCodex
Use when resolving Spec Kit open questions, NEEDS CLARIFICATION markers, or ambiguous feature requirements before planning.
Central agentic standards, guidelines, skills & bootstrap tooling for AI coding agents (Claude, Codex, Copilot).
This repository also configures its own agents. See what agent-compass tells them →
Skill Claude CodeCodex
Use when resolving Spec Kit open questions, NEEDS CLARIFICATION markers, or ambiguous feature requirements before planning.
Skill Claude CodeCodex
Use when creating or updating a Spec Kit project constitution, governance rules, principles, or non-negotiable constraints.
Skill Claude CodeCodex
Use after implementation to reconcile Spec Kit artifacts, code, tests, docs, and implementation status.
Skill Claude CodeCodex
Use when executing approved Spec Kit tasks after spec, plan, tasks, and checklists are ready.
Skill Claude CodeCodex
Use when creating a Spec Kit technical plan, research notes, data model, contracts, quickstart, or validation plan from an approved spec.
Skill Claude CodeCodex
Use when running a Spec Kit specify step to create or update a feature specification from a natural-language feature request.
Skill Claude CodeCodex
Use when generating Spec Kit implementation tasks from an approved plan and design artifacts.
Skill Claude CodeCodex
Drive and enforce the OpenSpec change lifecycle: resolve the one root, ask the status gate which artifact is next, keep proposal, delta specs, design and tasks coherent, verify before archiving, and fail the gate when a change is half-planned. Use when the project keeps specs under an openspec/ or docs/openspec/ root…
Skill Claude CodeCodex
Use when converting Spec Kit tasks.md items into deduplicated GitHub issues tied to the current repository.
Skill Claude CodeCodex
Split delivery work across the project's personas (profiles) and write the assignment plan as validated JSON. Use when the user asks who does what, to dispatch or assign tasks, to split work by profile or role, to build a work-breakdown per developer, or to produce an assignment brief per team member.
Skill Claude Code
Change verification gate. Analyzes code changes, checks documentation sync, and assesses change impact. Use when the user mentions change checks, doc sync, code review, pre-commit checks, or diff analysis. Auto-triggers for design-level changes and completed refactors.
Skill Claude Code
Module completeness gate. Scans the directory structure, detects missing documentation, and verifies code/doc sync. Use when the user mentions module verification, doc checks, structural completeness, README checks, or DESIGN checks. Auto-triggers when a new module is completed.
Skill Claude Code
Code quality gate. Detects complexity, duplicated code, naming-convention violations, function length, and other quality metrics. Use when the user mentions code quality, complexity checks, code smells, refactoring suggestions, lint checks, or coding standards. Auto-triggers for complex modules and completed refactors.
Skill Claude Code
Security verification gate. Scans code for vulnerabilities, detects dangerous patterns, and ensures security decisions are documented. Use when the user mentions security scans, vulnerability detection, security audits, code security, OWASP, injection detection, or sensitive-data leaks. Auto-triggers for new modules…
Skill Claude CodeCodex
Prove that a built screen matches its reference — a Figma export, a generated HTML preview, or a mockup image — with deterministic Playwright screenshots. Use when the user asks for visual regression, screenshot tests, pixel diffs, "does the build match the design", or when you change any user interface and must ship…
Skill Claude CodeCodex
Split a backlog of work-items across people or agents with clear separation of concern, minimal file overlap, explicit coordination seams, and a shareable assignment brief per owner. Use when dividing work across a team, deciding who does what, or parallelizing work so contributors don't collide.
Skill Claude CodeCodex
Use when the user asks to change how you communicate or how much you build — be terse, use fewer tokens, caveman mode, stop over-engineering, be lazy, YAGNI, simplest solution, action-first output, no preamble, or rewrite text so another agent cannot misread it. Installs the tracked style skill that matches the…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: