By outcomeeng
Prose craft for external prose (/writing-prose, /auditing-prose) and internal team docs (/writing-internal-docs, /auditing-internal-docs)
ALWAYS invoke this skill when auditing or reviewing internal team documents for cleanup. Use this whenever the user asks to review, audit, clean up, or check the writing of a Notion page, runbook, hiring rubric, scorecard, internal policy, decision record, design spec, competency model, onboarding guide, or any other doc that lives in a workspace and is read by colleagues. NEVER invoke for external-facing prose like READMEs, blog posts, web copy, customer release notes, or marketing material — use auditing-prose for those instead. NEVER invoke for responses to the user, including long research summaries delivered in chat, or for agent-facing instructions like SKILL.md and AGENTS.md.
ALWAYS invoke this skill when auditing reader-facing documents such as public docs, web pages, and product messages for outside readers like developers and customers. NEVER invoke for chat responses to the user (no matter how long), operational prose like code comments, commit messages, or agent-facing instructions like AGENTS.md and SKILL.md.
Catalog of anti-patterns and positive patterns for internal team documents (Notion pages, runbooks, scorecards, hiring rubrics, internal policies, decision records, design specs, competency models). Reference skill loaded by other internal-doc skills, not invoked directly. Use writing-internal-docs to write, or auditing-internal-docs to review.
Prose anti-patterns enforced across all skills. Loaded by other skills, not invoked directly.
ALWAYS invoke this skill when writing or editing internal team documents that live in a workspace: Notion pages, runbooks, hiring rubrics and scorecards, internal policies, decision records, design specs, competency models, onboarding guides, status pages, internal wiki content. Use this skill whenever the user is creating a doc intended for colleagues who already have context, not for strangers reading it for the first time. NEVER invoke for external-facing prose like READMEs, blog posts, web copy, customer release notes, or marketing material — use writing-prose for those instead. NEVER invoke for chat replies, commit messages, code comments, or agent-facing instructions like SKILL.md and AGENTS.md.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A combined Codex and Claude Code plugin marketplace for Outcome Engineering — the product engineering paradigm where a durable map of your product, maintained as a Spec Tree, serves as the authoritative source of truth for all implementation.
This repository publishes two plugin surfaces from the same source tree:
.claude-plugin packages for Claude Code plugins, commands, and agents.codex-plugin packages for Codex skill bundlesAGENTS.md is a symlink to CLAUDE.md, so Codex and Claude Code read the same repo-level instructions.
/bootstrapinterviews you about your product, then scaffolds a spec tree — the durable map that drives all implementation.

npm install -g @outcomeeng/spx
The spx CLI is the developer tool for Spec Tree maintenance and validation. Required by all engineering plugins.
claude plugin marketplace add outcomeeng/plugins
codex plugin marketplace add outcomeeng/plugins
Codex registers the marketplace source in the user's ~/.codex/config.toml and reads the shared .codex-plugin bundles from it.
# Spec Tree methodology (requires spx CLI)
claude plugin install spec-tree@outcomeeng
# Language plugins (install per product, require spx CLI)
claude plugin install python@outcomeeng
claude plugin install typescript@outcomeeng
# Optional plugins
claude plugin install prose@outcomeeng
claude plugin install develop@outcomeeng
After adding the marketplace, enable only the plugins a product needs in that repo's committed .codex/config.toml:
[plugins."spec-tree@outcomeeng"]
enabled = true
[plugins."develop@outcomeeng"]
enabled = true
Add language plugins in projects that use them:
[plugins."python@outcomeeng"]
enabled = true
[plugins."typescript@outcomeeng"]
enabled = true
Add domain plugins the same way:
[plugins."frontend@outcomeeng"]
enabled = true
[plugins."visual@outcomeeng"]
enabled = true
[plugins."hdl@outcomeeng"]
enabled = true
> /bootstrap # set up a new spec tree

> /author outcome for search # author a new outcome node
> /author PDR for auth policy # author a product decision
> /author ADR for caching strategy # author an architecture decision
> /apply # start the TDD flow
> /commit # commit with Conventional Commits
See the full tutorial for the complete workflow — from bootstrapping to handoffs.
claude plugin marketplace update outcomeeng
codex plugin marketplace upgrade outcomeeng
From this checkout, just push-marketplace wraps the Codex upgrade with cache
path preservation so active sessions with stale skill paths keep resolving for
seven days.
When working on the marketplace itself, every branch that changes a plugin's
distribution surface bumps that plugin's version exactly once. The just bump
recipe automates this — it detects which plugins changed under src/plugins/<name>/**
since origin/main, classifies each plugin's change pattern into a semver
segment, and updates the version field in every manifest each plugin owns
(.claude-plugin/plugin.json and, when present, .codex-plugin/plugin.json)
in lockstep:
just bump # auto-detect segment per plugin vs origin/main
just bump-dry # preview without writing
just bump-check # CI gate: exit non-zero if any changed plugin needs a bump
just bump origin/main minor # force minor for every changed plugin (warns on disagreements)
just bump main major # major bump vs local main (must be explicit)
npx claudepluginhub outcomeeng/plugins --plugin proseClaude Code plugin agents, commands, and skills
Legacy Outcome Engineering (spx/): PRDs, ADRs, capabilities, features, stories
Visual diagrams: /excalidrawing (Excalidraw JSON generation)
Python development workflow with testing, architecture, and coding skills
Prose craft: /writing-prose (always active), /reviewing-prose (on request)
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Build and maintain an LLM-curated personal knowledge base in your project — Andrej Karpathy's LLM Wiki pattern, designed to scale to thousands of pages without becoming a context bottleneck. Now with an optional compiled graph layer for typed, provenance-backed relationships.
Connect to Atlassian products including Jira and Confluence. Search and create issues, access documentation, manage sprints, and integrate your development workflow with Atlassian's collaboration tools.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.
20 modular skills for idiomatic Go — each under 225 lines, backed by 48 reference files, 8 automation scripts (all with --json, --limit, --force), and 4 asset templates. Covers error handling, naming, testing, concurrency, interfaces, generics, documentation, logging, performance, and more. Activates automatically with progressive disclosure and conditional cross-references.
Claude + Obsidian knowledge companion. Sets up a persistent, compounding wiki vault (Karpathy's LLM Wiki pattern). v1.7 "Compound Vault" + v1.8 methodology modes close 5 of 5 priority gaps from the May 2026 compass artifact. Ships: substrate alignment with kepano/obsidian-skills, default Obsidian CLI transport, hybrid retrieval (contextual prefix + BM25 + cosine rerank per Anthropic's Sept 2024 research), per-file advisory locking for multi-writer safety, pre-commit verifier agent, AND methodology modes (LYT / PARA / Zettelkasten / Generic) for first-class organizational support no other Claude+Obsidian competitor offers. v1.7.x audit closure: every BLOCKER + HIGH + MEDIUM + LOW finding from the v1.7.0 audit is CLOSED or DEFERRED-with-rationale. Optional DragonScale Memory extension (log folds, deterministic addresses, semantic tiling lint, boundary-first autoresearch).