claude

Query Claude Code session history via a DuckDB index over `~/.claude/projects/`. Use when asked about Claude Code activity ("how many tokens today?", "what did I work on this week?") or instead of reading, grepping, or jq-ing session transcripts. Not for codebase search, git log queries, or arbitrary databases.

Creating and optimizing Claude Code Skills including activation patterns, content structure, and development workflows. Use when creating new skills, converting memory files to skills, debugging skill activation, or understanding skill architecture and best practices.

Read and query Things 3 data (lists, todos, projects, tags, logbook). Not for writes. Use things:url to create/update, things:inbox for quick captures.

Triage and batch-implement Claude-tagged Things todos as PRs for the claude config repo, discover improvement candidates from session history, or watch open PRs to implement review feedback and close shipped todos. Use when the user wants to work on their Claude Code improvement backlog, process Things todos tagged claude-code, batch-implement configuration changes, mine session history for grounded config-change candidates (Discover mode), or watch this skill's open PRs for review feedback and merges (Watch mode).

Fetch, reply to, and resolve review threads on a GitHub pull request. Use when checking what review feedback needs addressing, whether threads are resolved, replying to review comments, or clearing AI-reviewer threads after acting on them.

Live tmux inbox for reviewing inbound pull requests across GitHub and GitLab. Use when reviewing multiple PRs, checking the review queue, batch reviews, or managing your review inbox. Pass --queue to spawn an already-ordered queue.

Review a pull request when requested by a peer. Use when reviewing PRs, providing code review feedback, or analyzing proposed changes. Supports GitHub and GitLab. Pass --triage to summarize a PR and estimate its review effort without reviewing it.

Harvest agent-tooling ideas from prominent developers.

Audit code comments for AI slop: restatement, narration/decision-log, self-praise, docstring-scope, and section-divider banners. Audits a diff (the comments a change introduced) or a whole repo, ranks by intrinsic complexity, fans out judging agents, and applies the trims to a fresh branch. Use when reviewing a branch or merge request, or sweeping a slop-heavy codebase.

Curate the writing plugin's trope ruleset by auditing wordlist entries against session history and surfacing candidate phrases. Use when refreshing trope detection, reviewing wordlist health, or mining sessions for new AI-writing patterns to add or stale rules to remove.

Create a pull request, merge request, or change request with proper formatting and content guidelines. Invoke when the user wants to create, open, or submit a PR, MR, or CR—including after committing changes.

Interacting with Linear issues, projects, and teams. Use when creating issues, updating issues, querying issues, managing projects, working on tasks, discussing backlogs, or any interaction with Linear.

Follow up on a PR/MR you reviewed: assess whether the author's fixes actually grasped each concern, find silently resolved or mechanically-fixed threads, and get a graded re-review call. Use after leaving review feedback to see if the author acted on it.

Drive tuicr, the terminal code-review TUI, as a live local review surface in tmux. Use when reviewing changes through tuicr: launching a session, seeding agent comments, collecting inline comments to act on, or watching comments arrive live. The shared core that review:self (inbound) and review:peer (outbound) delegate to. Triggers on "review in tuicr", "tuicr session", "open the diff in tuicr", "watch my comments".

Refining issues with technical context and structured details. Use when expanding a brief bug, feature, or refactor description into a detailed issue suitable for developers and AI agents.

Implement a feature or fix based on an issue. Use when given an issue URL to work on, or when implementing changes described in a tracked issue. Supports GitHub, Linear, and GitLab.

Signal that you are stepping away so the agent keeps working without stalling on prompts. Drives the current task to completion, batches reversible decisions behind safe defaults, raises only irreversible ones via push, and defers bandwidth- or power-heavy work when tethered, on battery, or lid-closed. Use via /afk or /afk <duration>.

Monitor GitHub Actions runs and extract failure diagnostics. Use when watching PR CI, branch builds, or specific workflow runs.

Investigate GitLab CI pipeline failures and extract diagnostics. Use when watching MR CI, branch builds, or specific pipelines.

Quick captures to the Things 3 inbox. Not for reads (things:jxa), scheduled tasks, updates, or projects (things:url).

Create and validate Mermaid diagrams. Use when creating, editing, or reviewing Mermaid diagrams in markdown files.

Tmux session, window, and pane management. Use when capturing output, sending keys, opening processes in panes, or checking notifications.

Daily routines for a corporate software engineering job. Start mode triages the inbound review queue, your own open PRs, and the issue tracker's plan for today. End mode clears the outbox, surfaces review debt, sweeps worktrees for unpushed work, and tidies tracker state for tomorrow. Use via /job, /job start, /job end, or /job setup.

Follow up on review feedback you received on a PR/MR: check what review comments need responses, investigate how threads were resolved (including silent resolves), and draft replies. With --auto, autonomously triage AI-reviewer (bot) threads and loop until the reviewer is satisfied, clearing a bot review hands-off.

Orchestrating Claude Code agent teams. Use when creating teams, spawning teammates, assigning tasks, configuring teammate modes, or setting up team quality gate hooks.

Resolving git merge conflicts during rebase, merge, or cherry-pick. Use when conflicts arise. Can also drive the operation to completion and push when asked (e.g. "fix conflicts and push").

Working with GitLab merge requests via glab. Use when creating, updating, reviewing, or merging MRs.

Monitor a PR's CI, fix trivial failures, and self-cancel when green; --merge drives to merged, --reviews hands off to AI-review triage.

Update a pull request or merge request body to reflect the current state of changes. Use when a PR/MR has evolved through additional commits and the body needs to reflect what will be merged.

Create, update, and manage Things 3 tasks and projects. Not for reads — use things:jxa to query data. For simple inbox captures, use things:inbox.

Review a document through specialized lenses using parallel agents. Use when reviewing documentation, blog posts, READMEs, proposals, or any prose for quality issues across content, style, and embedded artifacts.

Measure Bun test coverage and close gaps on a specific file. Use when adding or editing tests, when asked about coverage, or when the PostToolUse coverage hook reports uncovered lines.

Score a single input for AI writing patterns on a per-1000-word basis. Use to measure trope density across vocabulary, marketing verbs, and structural patterns, compare two drafts, or score code comments separately, without rewriting. Reports hit counts and density per category, never gates.

Shorthand to make Claude convert a wall of prose into structured questions via the AskUserQuestion tool. Invoke with /ask.

Survey a project for ways to adopt newer language and major-dependency features that did not exist when the code was written, replacing boilerplate with modern idioms. Use when the user wants to modernize a codebase, adopt new language features, retire dependencies the stdlib now subsumes, or catch a project up to current framework APIs.

Bun runtime patterns. Use when running bun commands, working with package.json/bun.lock, writing TypeScript scripts under Bun, or developing Claude Code plugins.

Configure, create, or troubleshoot Claude Code hooks (PreToolUse, PostToolUse, UserPromptSubmit), debug hook failures, or set up any automation within Claude Code. Examples include "I want to run tests before every file edit", "My hook isn't firing", "1 out of 2 hooks ran", or "How do I create a hook that formats JSON output with jq?"

Authoring and validating GitLab CI configuration. Use when editing `.gitlab-ci.yml`, structuring pipeline stages and jobs, or running `glab ci lint`. For pipeline/job failure investigation, use gitlab:ci-monitor.

Comprehensive interview for planning new features or changes. Use when the user wants thorough upfront planning before implementation, or when starting work on a complex feature.

Execute JXA scripts scoped to a macOS application. Use when running JXA expressions or script files against an app via osascript. Validates Application() scope via AST.