Skip to main content

By agent · Codex

Codex skills: 1008 worth installing

Skills in the open SKILL.md format that Codex reads. Each one has a plain-language summary and the command to install it.

Installing a skill in Codex

$ npx skills add <author>/<skill> --agent codex

Replace the placeholder with the author and skill name from the skill's page. Start a new Codex session afterwards so the skill is picked up.

Skills for Codex

All skills

Develops Base44 apps remotely in the cloud sandbox, authoring functions, actors, entities, agents and connectors without a local checkout.

DevelopmentGitHub stars: 91Installs: 5.4k

BMAD

BMad Code Org

Guides BMad projects by answering questions, recommending the next installed skill, and handling setup, migration, initiative changes and installation checks.

ProductivityGitHub stars: 53.7kInstalls: 5.6k

BMAD Code Review

BMad Code Org

Reviews code changes with BMAD by running independent reviewers in parallel, then triaging and presenting their findings.

DevelopmentGitHub stars: 53.7kInstalls: 5.5k

BMAD Agent PM

BMad Code Org

Guides BMAD product-management sessions through requirements discovery, stakeholder alignment and PRD creation.

ProductivityGitHub stars: 53.7kInstalls: 5.5k

Guides Dart API documentation with Effective Dart rules for doc comments, linking, Markdown formatting and analysis verification.

DevelopmentGitHub stars: 3kInstalls: 5.3k

BMAD Brainstorming

BMad Code Org

Facilitates BMad brainstorming sessions with selectable stances, creative techniques, persistent session logs and optional convergence.

ProductivityGitHub stars: 53.7kInstalls: 5.5k

BMAD Agent Dev

BMad Code Org

Implements approved BMAD stories and code changes with test-first development, acceptance-criteria checks and verified delivery.

DevelopmentGitHub stars: 53.7kInstalls: 5.4k

BMAD Agent Analyst

BMad Code Org

Guides the BMAD Method's Mary business analyst through market research, competitive analysis and requirements work from a project-aware menu.

ResearchGitHub stars: 53.7kInstalls: 5.4k

BMAD Build

BMad Code Org

Turns delegated BMAD implementation work into code through a rendered workflow with review and verification for features, stories and bug fixes.

DevelopmentGitHub stars: 53.7kInstalls: 5.4k

BMAD Product Brief

BMad Code Org

Creates, updates and validates product briefs through guided discovery, source review, decision logging and right-sized documentation.

ProductivityGitHub stars: 53.7kInstalls: 5.3k

BMAD Agent UX Designer

BMad Code OrgAudit warning

Acts as BMAD's UX designer, translating user needs into interaction designs and UX specifications through a configured design persona and workflow.

DesignGitHub stars: 53.7kInstalls: 5.3k

BMAD Architecture

BMad Code Org

Creates concise BMad architecture spines that record cross-component invariants, boundaries, state ownership and deferred decisions from ideas, specs or codebases.

DevelopmentGitHub stars: 53.7kInstalls: 5.3k

BMAD Spec

BMad Code OrgAudit warning

Distils ideas, briefs, PRDs and mixed notes into a lean BMad requirements specification with stable capabilities, constraints and validation.

ProductivityGitHub stars: 53.7kInstalls: 5.3k

BMAD PRD

BMad Code Org

Creates, updates and validates product requirements documents in the BMad Method, with guided discovery, research, review and finalisation.

ProductivityGitHub stars: 53.7kInstalls: 5.3k

BMAD Correct Course

BMad Code Org

Assesses significant sprint changes in the BMAD Method across requirements, architecture and UX, then produces an actionable change proposal.

ProductivityGitHub stars: 53.7kInstalls: 5.2k

Generates automated API and end-to-end tests for implemented features in the BMAD Method, using the project's existing test framework and patterns.

DevelopmentGitHub stars: 53.7kInstalls: 5.2k

BMAD Retrospective

BMad Code Org

Reviews a finished BMAD epic against its plans, diffs and commits, then produces sourced findings, action items and an acceptance decision.

ProductivityGitHub stars: 53.7kInstalls: 5.2k

BMAD UX

BMad Code Org

Captures a product's UX vision in BMad through paired design and experience specifications, guided discovery, user journeys and validation.

DesignGitHub stars: 53.7kInstalls: 5.1k

BMAD Review

BMad Code OrgAudit warning

Runs configurable BMAD review lenses over code or documents, combining adversarial, edge-case, verification, structure and prose findings into a triaged report.

DevelopmentGitHub stars: 53.7kInstalls: 5k

Maintains a repository's AGENTS.md guidance by setting up, adopting, refreshing or auditing instructions and recording agent mistakes as pitfalls.

DevelopmentGitHub stars: 53.7kInstalls: 5k

BMAD Forge Idea

BMad Code Org

Pressure-tests a developing idea through focused questions and contrasting personas, then records its outcome and optional planning brief in BMad.

ProductivityGitHub stars: 53.7kInstalls: 5k

BMAD Deep Recon

BMad Code Org

Researches decisions through drafted prompts, parallel web searches or cited summaries of existing reports in the BMad workflow.

ResearchGitHub stars: 53.7kInstalls: 4.9k

BMAD Build Auto

BMad Code Org

Runs one unattended development-loop iteration through BMad, with optional route and review selections.

DevelopmentGitHub stars: 53.7kInstalls: 4.9k

Drafts and revises pull request descriptions using the repository's template, verified branch facts, validation results and reviewer guidance.

DevelopmentGitHub stars: 599Installs: 4.7k

Questions

Codex and skills.

Do Claude skills work in Codex?

Often. A skill is a folder of instructions and scripts in the open SKILL.md format, which Codex reads. A skill that depends on tools only one agent provides will not carry over, so read the skill's source before you rely on it.

How do I install a skill in Codex?

Each skill page has an install command. Run it in a terminal, then start a new Codex session so the skill is picked up.

What is the difference between a skill and an instructions file such as AGENTS.md?

An instructions file is loaded in every session and describes the project. A skill is loaded only when a task calls for it, so it can be long and specific without costing context on every turn.

Other agents