This is a portable engineering toolkit, cross-harness AI agent skills library of AI agent workflows that helps teams turn ideas into well-scoped, verifiable software changes, across Claude Code, Copilot, Codex, Cursor, and other coding environments, with a cross-platform installer.
This repository also configures its own agents. See what zen-agent-skills tells them →
Skill Claude CodeCodex
Turns the current session's context into a self-contained, execution-ready brief that a fresh agent session or subagent can run without this conversation. Use when work moves to another agent or a new session and the instructions should be written down first: "draft a prompt for a new session", "craft a prompt I can…
Skill Claude CodeCodex
Answers questions about agent sessions across every project at once by reading the local observatory store: which sessions exist and are running, what each cost, which skills were used, and what failed. Where the harness allows, it can rename a session or message another one; it never starts, resumes, interrupts, or…
Skill Claude CodeCodex
Writes new Markdown documentation grounded in the repository's actual code: READMEs for technical or non-technical readers, architecture docs with Mermaid diagrams, setup, deployment, and contributing guides, design docs, and decision records. Use when a document does not exist yet: "create a README", "write an…
Skill Claude CodeCodex
Revises existing Markdown documentation in place without flattening its voice or rewriting more than asked: reconciles a doc with the current code, fixes cross-document links after files move, and keeps a doc set consistent. Use when documentation already exists and needs changing: "update the docs to reflect this"…
Skill Claude CodeCodex
Detects documentation drift by checking prose claims against repository facts, after code has moved or when a doc set needs auditing for stale claims. Classifies each document as current-state, contract (report-only, human-owned), or ledger (skipped), and reports each finding with a stable id, the grounding fact, and…
Skill Claude CodeCodex
Dispatches a batch of independent, already-scoped task files (bugs, chores, small features) to parallel agents, each in its own git worktree, then verifies every result independently before anything is called mergeable. Use when there are two or more independently fixable items to action: "fix these bugs", "work…
Skill Claude CodeCodex
Reviews a code change against an explicit house rubric and severities and writes a severity-ranked markdown review, without editing or committing anything. Resolves every finding's quoted evidence against the real file and drops any that does not resolve, then reports blocker to nit with file:line, the quote, a stable…
Skill Claude CodeCodex
Packages a project's current state into a handoff for a person (a business partner, client, or teammate) as a document or short message in plain language tuned to the reader, with internals and costs redacted for a client. Use to bring someone up to speed, ask a collaborator to decide something, or prep a meeting…
Skill Claude CodeCodex
Scaffolds a portable, agent-ready work-tracking system into the current repository: AGENTS.md with an agent reading protocol, a .tasks/ directory of atomic task files with validate.py, and by tier a ROADMAP.md, a CHANGELOG.md, and Claude Code, Cursor, and Copilot pointers. Picks a lite, standard, or team tier so a…
Skill Claude CodeCodex
Authors atomic, agent-ready task files for a .tasks/ system: honest touchedfiles, a real parent, resolved dependson, and a mechanically verifiable acceptance command, self-checked with validate.py and given collision-safe ids. Elicits and decomposes before writing. Use for "write a task for X", "add this to the…
Skill Claude CodeCodex
Drafts a pull request description and a matching changelog entry from a branch's diff, in the target repository's own changelog format, without touching GitHub. Finds the commit range against the default branch, surveys the change, and references the work item when a .tasks/ system exists; it prints the gh command…
Skill Claude CodeCodex
Lands the work in one or more git worktrees (left by parallel agents or by fix-batch) in the main working tree deliberately, without blind commits or merges: finds each worktree's real base, catches untracked files, binary assets, and overlaps, and applies one worktree at a time. Use for "reconcile the worktrees"…
Skill Claude CodeCodex
Decides how much review a change has earned before reviewing it: selects quick, standard, or deep from stated signals (changed lines, directory spread, trust-boundary flags, blast radius, docs-only scope), reports the signal table and the rule that fired, then runs house-review at that depth. Use when a change is…
Skill Claude CodeCodex
Use when turning a raw idea, feature request, or bug into a persistent behavioral spec before any tasks or code. Drafts a scenario-based spec (problem, goals, non-goals, constraints, Given/When/Then scenarios with stable S-NNN ids, proposed surface, open questions), composes the spec-quality lens to self-check and…
Skill Claude CodeCodex
Use when closing a spec-driven feature or issue, or auditing whether an implementation actually matches its spec. Given a spec path, emits a section-by-section conformance matrix mapping every scenario and proposed surface element to conformed (with file:line or test evidence), diverged (what, why, and both sides), or…
Skill Claude CodeCodex
Use before implementation when a spec and its task decomposition must be checked for coding readiness. Applies a deterministic go/no-go gate for missing inputs, unresolved spec blockers, incomplete ordered task lists, spec-task contradictions, task-to-scenario traceability, scenario-to-test layer mapping…
Skill Claude CodeCodex
Use when drafting, reviewing, repairing, or validating scenario-based specs, product specs, implementation specs, Proposed Surface sections, Given/When/Then scenarios, Open Questions, or spec-quality checklists. A quality lens for spec well-formedness: keeps specs contract-level, observable, non-redundant, grounded in…
Skill Claude CodeCodex
Derives runnable tests from an approved spec's Given/When/Then scenarios and a task's acceptance criteria, one faithful test per scenario tagged with its S-NNN id, in the repository's own test framework. Composes the test-quality lens for the lowest faithful layer and an exact-outcome oracle, writes characterization…
Skill Claude CodeCodex
Use when creating, modifying, reviewing, or deciding whether to add tests, fixtures, mocks, integration tests, end-to-end tests, smoke tests, or test plans in any codebase or language, including regression tests for bug fixes. Chooses the test layer and oracle: guides agents to pick the right test layer, avoid…
Skill Claude CodeCodex
Independently verifies an implementation against its approved spec and its task's acceptance criteria before it is reconciled or landed. Runs the declared commands and records each exact outcome, composes spec-conformance so a contract divergence fails the run even when every command passes, maps each criterion to met…
Skill Claude CodeCodex
Bootstrap a new or bare repository with a sane, stack-matched baseline, then hand off to init-worktracking for work tracking. Detects the language and package manager, then lays down .gitignore, .editorconfig, a linter/formatter config, a LICENSE, and a README stub, without clobbering anything that already exists. Use…
Skill Claude CodeCodex
Establishes the cause of a defect, test failure, flaky test, or unexpected behavior before anyone writes a task or a fix: reproduces it first, localizes the failure to the boundary where behavior diverges, tests one hypothesis at a time, and returns a deterministic verdict, so a run that found nothing never reads as…
Skill Claude CodeCodex
Plans and logs a season of project work on a shared quest tracker page: phases of steps that each say what to do, why, and what closes it, with line ticks, facts, a session log, and a quest chain, saved in the page's own database. Starts a tracker from the bundled template, works a session against one, and has agent…
Skill Claude CodeCodex
Answers "where are we?" about one repository from its own records rather than memory: what waits on the person (an open question no decision answers, or work held for a human eye), what is blocked on an unfinished dependency, and what is in flight in worktrees and unmerged branches, with an evidence tier on every…
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: