Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Audit and improve CLAUDE.md files in repositories. Use when user asks to check, audit, update, improve, or fix CLAUDE.md files. Scans for all CLAUDE.md files, evaluates quality against templates, outputs quality report, then makes targeted updates. Also use when the user mentions "CLAUDE.md maintenance" or "project memory optimization".

Evaluate changes from upstream sources (gsd-pi and pi-mono) for potential integration into kata-mono's CLI app. Use when the user mentions upstream changes, syncing from gsd-pi or pi-mono, checking what's new upstream, evaluating features to cherry-pick, comparing gsd-pi extensions with kata's, checking for pi-coding-agent updates, or asks "what's new in gsd/pi-mono". Also use when the user says "upstream eval", "sync check", "feature delta", "what should we pull in", or references /Volumes/EVO/kata/gsd-pi or /Volumes/EVO/kata/pi-mono in a change-evaluation context.

Structural and semantic codebase intelligence with persistent memory for AI coding agents. Indexes TypeScript and Python repositories into a SQLite knowledge graph with optional vector embeddings, then exposes graph queries, semantic search, pattern search, fuzzy symbol lookup, and durable memory op

Run user acceptance testing for Kata Desktop, the Electron app at apps/desktop. Covers launching the app with CDP, connecting agent-browser, walking through UI flows, capturing screenshots, and writing structured UAT reports. Use this skill whenever the user mentions "UAT", "acceptance test", "validate the desktop app", "test the electron app", "walk through the app", "dogfood desktop", "screenshot the app", "verify the milestone", or any request to visually inspect or interactively test Kata Desktop. Also use when finishing a milestone or slice that includes Kata Desktop UI work, or when creating a UAT report for a PR.

Use when the user wants to execute a standalone Kata issue, one-off issue plan, backlog issue, or isolated implementation plan created by kata-plan-issue. This skill selects or retrieves an open issue from the backend, confirms the issue with the user, then executes its approved plan with fresh subagents and two-stage review per task.

Use when the user asks for Kata project status, milestone progress, task progress, or execution state.

Use when the user asks to install Kata, set up Kata CLI, connect Kata to an agent harness, or check initial Kata configuration.

Keep all documentation current when code changes land. Covers the docs site (apps/online-docs) and per-app internal docs (README, AGENTS.md, reference files, changelogs). Use after making changes to any app or package that affect user-facing behavior, configuration, architecture, or developer workflows. Also use when adding/removing/renaming commands, skills, extensions, or preference fields. Fires on phrases like "update docs", "sync documentation", or proactively when implementation changes land.

Use this skill when releasing Kata Desktop, Kata CLI, Kata Orchestrator, Kata Context, or Symphony, bumping versions, updating changelogs, or creating release PRs. Triggers include "release", "bump version", "publish", "create release PR", "ship it", "cut a release".

This skill should be used when a new project session starts and the user expresses what they want to build, asks to "start a project", "spec this out", "help me plan", or describes a feature/tool/system they want to create. Guides structured intent capture through goal, constraints, architecture, acceptance criteria, tasks, and non-goals.

Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Create a well-formed git commit from current changes using session history for rationale and summary; use when asked to commit, prepare a commit message, or finalize staged work.

Investigate stuck runs and execution failures by tracing Symphony and Codex logs with issue/session identifiers; use when runs stall, retry repeatedly, or fail unexpectedly.

Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions. Trigger words include: 'fix CI', 'debug GitHub Actions', 'gh pr checks', 'CI is red', 'GitHub Actions failed', 'fix the build', and similar phrases indicating a need to investigate and resolve CI failures in a GitHub-hosted repository.

Land a PR by monitoring conflicts, resolving them, waiting for checks, and squash-merging when green; use when asked to land, merge, or shepherd a PR to completion.

Use Symphony's `linear_graphql` client tool for raw Linear GraphQL operations such as comment editing and upload flows.

Pull latest origin/<base-branch> into the current local branch and resolve merge conflicts (aka update-branch). Use when Codex needs to sync a feature branch with origin, perform a merge-based update (not rebase), and guide conflict resolution best practices.

Push current branch changes to origin and create or update the corresponding pull request (with the correct base branch); use when asked to push, publish updates, or create pull request.

Use this skill when the user wants to prove, test, UAT, validate, capture evidence for, or clean up backend integrations for either Kata CLI or the Symphony runtime against real GitHub Projects v2 or Linear instances. Use it for Kata CLI operation coverage, Symphony helper operation coverage, backend health checks, provider proof links, evidence bundles, cleanup of prior UAT runs, and keeping the UAT harness current when CLI or Symphony runtime contracts change.