claude-marketplace

Install and maintain a "model routing" section in a project's CLAUDE.md that teaches Claude Code which model to use for which work — including delegating to GPT models via the Codex CLI. Use when the user wants to add model-picking guidance to CLAUDE.md, set up Codex/GPT as a delegation target or fallback, route subagent and workflow tasks across models by cost/intelligence/taste, or says "add model routing", "use codex as fallback", or "set up multi-model workflows".

The non-interactive `rg`/`fd`/`rga` → `fzf --filter` pipeline — scan with a regex tool, then fuzzy-rank the lines without opening a TUI. Use when you want fuzzy (not regex) matching of file paths, code lines, or document/PDF text from a script or one-shot command, when reproducing the fuzzy-search MCP's behavior on the bare command line, or when piping fuzzy-ranked results into xargs/an editor. This is the CLI technique; for the INTERACTIVE picker use the fzf skill, and for tool-call ergonomics use the fuzzy-search MCP server.

Configure the Claude Code status line with VCS-aware scripts showing git branch, jj change ID, bookmarks, context usage, and costs. Use when setting up a statusline, customizing the status bar, adding git or jj info to the prompt, configuring statusLine in settings.json, or troubleshooting statusline scripts.

Create a repo-local Agent Skill following the open agentskills.io specification. Use when the user wants to create, scaffold, or initialize a new skill in the current repo, mentions a 'repo-local skill' or the agentskills.io spec, or needs a spec-compliant SKILL.md placed under .agents/skills/ (or .claude/skills/). Scaffolds the directory, writes valid name/description frontmatter, and validates with skills-ref.

The pytest testing framework and its Python-specific idioms — the things that set pytest apart from unittest and generic testing. Use when writing, running, debugging, or configuring Python tests with pytest — plain `assert` (rewritten for rich introspection), dependency-injected `@pytest.fixture` (scopes, autouse, yield-teardown, conftest sharing), table-driven `@pytest.mark.parametrize`, `skip`/`skipif`/`xfail` and custom markers, `pytest.raises`/`warns`/`approx`, builtin fixtures (`tmp_path`, `monkeypatch`, `capsys`, `caplog`), the CLI (`-k`/`-m`/`-x`/`--lf`/`--pdb`/node ids), config (`pyproject.toml`/`pytest.ini`/`pytest.toml`, `addopts`, `testpaths`, `--import-mode`), and plugins/hooks (`conftest.py`, `pytest_addoption`, xdist/cov/asyncio/mock). Includes inline `(pytest N.M+)` version annotations + a version-features lookup. The pytest framework — not unittest, not tox/nox runners, not coverage.

Compact the current conversation into a handoff document for another agent to pick up. Use when ending a session with unfinished work, transferring context to a fresh agent or a new session, approaching context limits, or when the user says "handoff", "write a handoff doc", or "continue this in another session".

Managing Linear issues, projects, and teams. Use when working with Linear tasks, creating issues, updating status, querying projects, or managing team workflows.

Command-line fuzzy finder for interactive filtering of any list. Use when interactively selecting files, searching command history (CTRL-R), creating selection interfaces in scripts, building interactive menus, or integrating fuzzy search with tools like ripgrep, fd, and git. Triggers on mentions of fzf, fuzzy finder, ** completion, interactive filtering, or shell keybindings CTRL-T/CTRL-R/ALT-C.

uv — Astral's single, fast (Rust) binary that replaces pip, pip-tools, pipx, pyenv, poetry/pdm, and virtualenv for Python packaging and project management. Use when managing a Python project with `pyproject.toml` + a universal `uv.lock` (`uv init/add/remove/sync/lock/run`), running PEP 723 inline-metadata scripts (`uv run script.py`), running one-off tools with `uvx` or installing them with `uv tool install`, installing/pinning Python versions (`uv python install/pin/list`), using `uv pip` as a faster drop-in pip/pip-tools replacement, creating venvs (`uv venv`), or building/publishing packages (`uv build`/`uv publish`). Triggers on mentions of uv, the `uv` command, `uvx`, `uv.lock`, `tool.uv`, `uv add/sync/run`, PEP 723 `# /// script` blocks, or "fast Python package manager". This is the **uv CLI by Astral** — NOT ruff (linter/formatter) or ty (type checker), which are separate Astral tools, and NOT poetry/pip/pyenv themselves.

The k3s lightweight CNCF-certified Kubernetes distribution shipped as a single binary that bundles containerd, flannel, CoreDNS, Traefik, ServiceLB (Klipper), local-path-provisioner, and metrics-server. Use when installing or running a lightweight Kubernetes cluster, bootstrapping a single-node or HA k3s control plane, joining agent (worker) nodes, choosing embedded etcd vs an external SQL datastore, managing cluster/agent tokens, rotating certificates, taking or restoring etcd snapshots, enabling secrets encryption, doing airgap or private-registry installs, disabling bundled components (Traefik/ServiceLB/CoreDNS), or upgrading k3s. Triggers on mentions of k3s, `k3s server`, `k3s agent`, the `curl -sfL https://get.k3s.io | sh -` installer, `/etc/rancher/k3s/`, or Rancher's lightweight Kubernetes. This is the k3s CLI/distribution, NOT generic Kubernetes/kubectl.

Author and operate Kubernetes workloads with kubectl and YAML manifests — the kubectl CLI plus core resource authoring (Pods, Deployments, StatefulSets, Jobs/CronJobs, Services, Ingress, ConfigMaps/Secrets, PV/PVC, RBAC). Use when writing or debugging Kubernetes manifests, running kubectl (get/describe/apply/diff/rollout/logs/exec/debug/port-forward), inspecting a failing pod (CrashLoopBackOff/ImagePullBackOff/Pending/OOMKilled), managing contexts/namespaces, generating manifests with --dry-run=client, checking permissions with auth can-i, or doing server-side apply. Triggers on kubectl, kube, k8s, Pod/Deployment/Service/Ingress/ConfigMap YAML, kubeconfig, kustomize. This is the kubectl + resource/manifest layer — for installing or running a lightweight cluster use the k3s skill; cluster install/admin, client libraries, cloud-provider specifics, and Helm are out of scope.

Invoke deep sequential thinking for complex problem-solving. Use when the user says 'use ultrathink', 'ultrathink', or when tackling problems that require careful step-by-step reasoning, planning, hypothesis generation, or multi-step analysis.

PostgreSQL server administration & operations — DBA work on a running cluster. Use when configuring a server (postgresql.conf, ALTER SYSTEM, reload vs restart, GUCs), managing roles & privileges (CREATE ROLE, GRANT/REVOKE, DEFAULT PRIVILEGES, predefined roles, membership), authentication (pg_hba.conf, scram-sha-256, peer/cert/ldap), backups & restore (pg_dump/pg_dumpall/pg_restore, pg_basebackup, incremental backups, PITR), replication & HA (streaming, logical publications/subscriptions, replication slots, failover, synchronous commit), pg_upgrade, or monitoring (pg_stat_activity, pg_stat_io, pg_stat_progress_*, log settings, checkpoints, connections). Server-side ops — for the psql client use `psql`, for SQL & query syntax `postgres-sql`, for query tuning & autovacuum mechanics `postgres-performance`, for contrib modules `postgres-extensions`. Features added in a release are tagged inline like `(pg16+)`; untagged items are bedrock (PostgreSQL 9.x or earlier). Verify with `SHOW server_version`.

PostgreSQL extension management and the bundled `contrib` catalog. Use when running `CREATE EXTENSION` / `ALTER EXTENSION ... UPDATE` / `DROP EXTENSION`, listing extensions (`\dx`, `pg_available_extensions`), or choosing which contrib module enables a feature — FDWs (`postgres_fdw`, `dblink`, `file_fdw`), trigram/fuzzy search (`pg_trgm`, `fuzzystrmatch`), crypto/UUIDs (`pgcrypto`, `gen_random_uuid`), types `hstore`/`ltree`/`citext`/`cube`, query stats (`pg_stat_statements`, `auto_explain`), index helpers (`btree_gin`, `btree_gist`, `bloom`), or storage forensics (`pageinspect`, `amcheck`, `pg_surgery`). Also covers trusted (non-superuser) extensions, `shared_preload_libraries`, and `CASCADE`. Disambiguation — extension management plus the contrib catalog; use psql for the client, postgres-sql for core SQL, postgres-performance for tuning *with* these extensions, postgres-admin for server config. Version notes are inline `(pgNN+)`; bedrock modules are unannotated, see references/version-features.md.

PostgreSQL query & performance tuning — reading EXPLAIN / EXPLAIN ANALYZE plans, choosing and designing indexes, fixing planner row-estimate errors with statistics, and diagnosing MVCC bloat / VACUUM. Use when a query is slow, when reading an EXPLAIN/EXPLAIN ANALYZE plan (seq vs index vs bitmap scan, nested-loop/hash/merge join, "Rows Removed by Filter", estimate-vs-actual skew), when choosing an index type or columns to index (btree/GIN/GiST/SP-GiST/BRIN/hash, multicolumn/partial/expression/covering), when ANALYZE/extended statistics give bad estimates, when tables bloat or autovacuum lags, or when tuning planner GUCs (random_page_cost, work_mem, effective_cache_size, enable_*). This is performance tuning — for the psql client see psql, for writing SQL/DDL see postgres-sql, for server config/replication & autovacuum GUC tuning see postgres-admin, for installing perf extensions (pg_stat_statements, auto_explain) see postgres-extensions. Newer features are tagged inline (pgNN+); PG18 is stable, PG19 is beta.

The PostgreSQL SQL dialect and data types that set Postgres apart from generic SQL. Use when writing or debugging Postgres-specific SQL — upsert (`INSERT ... ON CONFLICT DO UPDATE/NOTHING`), `RETURNING` (incl. `OLD`/`NEW`), `MERGE`, CTEs (`WITH`/`MATERIALIZED`/recursive/data-modifying), window frames, `GROUPING SETS`/`CUBE`/`ROLLUP`, `DISTINCT ON`, `LATERAL`, `FILTER`, generated & identity columns, declarative partitioning, `jsonb` operators + jsonpath + SQL/JSON (`JSON_TABLE`/`IS JSON`), arrays, ranges & multiranges, composite/enum/domain types, and functions like `uuidv7()`/`gen_random_uuid()`/`ANY_VALUE()`. Triggers on `ON CONFLICT`, `MERGE`, `jsonb`, `@>`/`->>`, `GENERATED ... AS IDENTITY`, `PARTITION BY`. Disambiguation — this is the SQL dialect & types; for the `psql` client use psql, for tuning/EXPLAIN/indexes use postgres-performance, for config/auth/backup use postgres-admin, for contrib use postgres-extensions. Features carry inline `(pgNN+)` tags; unlisted = long-standing (bedrock since 9.x).

psql — PostgreSQL's interactive terminal client and SQL script runner (the `psql` command). Covers connection strings/URIs, `~/.pgpass` & service files, invocation flags (`-c`/`-f`/`-1`/`--csv`/`-qAtX`/`-v`), the `\d`-family inspection meta-commands, `\copy`, `\watch`, `\gexec`/`\gset`, `\if` scripting, `\pset` output formats & `\x`, variables & `:'…'` interpolation, `.psqlrc`, exit codes/`ON_ERROR_STOP`, and extended-protocol/pipeline commands (`\bind`/`\parse`/`\startpipeline`). Use when connecting to or inspecting a Postgres database from a terminal, running SQL scripts or one-off queries in shells/CI/agents, formatting or capturing query output, or fixing psql connection/auth problems. Includes inline (pgNN+) version annotations (pg14–pg19). This is the psql CLIENT only — for SQL syntax/data types use postgres-sql; for EXPLAIN/indexes/vacuum use postgres-performance; for server config/auth/backup/replication use postgres-admin; for CREATE EXTENSION use postgres-extensions.

jq — the command-line JSON processor and its filter language for slicing, filtering, transforming, and reshaping JSON. Use when extracting fields or nested paths from JSON on the command line, filtering/`select`-ing arrays of objects, reshaping or building new JSON objects, emitting CSV/TSV with `@csv`/`@tsv`, getting raw unquoted strings for shell capture with `jq -r`, slurping (`-s`) and aggregating JSON (`add`, `group_by`, averages), parsing API responses, NDJSON pipelines (`-c`), or injecting shell values safely with `--arg`/`--argjson`/`$ENV`. Triggers on mentions of jq, the `jq` command, the jq filter language, `jq -r`, `.foo`/`.[]` filters, `@csv`/`@tsv`, slurp/`fromjson`, or "filter/transform JSON in the terminal". This is the jq CLI and filter language, NOT a generic JSON tutorial and NOT jq language bindings (jaq, gojq, PyPI `jq` — defer those to their own docs).

DuckDB — the in-process, columnar, vectorized OLAP SQL engine shipped as a single zero-dependency binary ("SQLite for analytics"). Use when querying Parquet/CSV/JSON files directly with SQL and no load step, doing in-process analytical/OLAP queries and aggregations, using the `duckdb` CLI shell (interactive REPL or `-c`/`-json` one-shots for scripts and agents), writing "friendly SQL" (FROM-first, `SELECT * EXCLUDE`, `GROUP BY ALL`, `SUMMARIZE`, `PIVOT`), converting CSV↔Parquet↔JSON with `COPY`, reading remote data over HTTP/S3 via the httpfs extension, or `ATTACH`-ing a live Postgres/MySQL/SQLite database. Triggers on mentions of duckdb, the `duckdb` command, `.duckdb` files, replacement scans, `read_parquet`/`read_csv`, httpfs/`CREATE SECRET`, or "embedded analytics database". This is the DuckDB CLI and SQL dialect, NOT a generic SQL tutorial and NOT the DuckDB client libraries (Python/Node/Java — defer those to duckdb.org).

Jujutsu (jj) version control system - a Git-compatible VCS with novel features. Use when working with jj repositories, managing stacked/dependent commits, needing automatic rebasing with first-class conflict handling, using revsets to select commits, or wanting enhanced Git workflows. Triggers on mentions of 'jj', 'jujutsu', change IDs, operation log, or jj-specific commands.

Create and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.

Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, properties, and other Obsidian-specific syntax. Use when working with .md files in Obsidian, or when the user mentions wikilinks, callouts, frontmatter, tags, embeds, or Obsidian notes.

ansible-core — the agentless, push-based, SSH automation engine that runs idempotent tasks from declarative YAML playbooks. Use when writing or running Ansible playbooks, issuing ad-hoc commands (`ansible -m`), managing INI/YAML inventory, authoring roles, installing collections with `ansible-galaxy`, encrypting secrets with Ansible Vault, tuning `ansible.cfg`, looking up module docs with `ansible-doc`, or provisioning/configuring hosts idempotently over SSH. Covers the ten `ansible*` CLIs, FQCN (`ansible.builtin.*`), become/privilege escalation, inventory patterns, variable precedence, check/diff mode, and strategies. Triggers on mentions of ansible, ansible-playbook, ansible-galaxy, ansible-vault, ansible-inventory, ansible.cfg, playbooks, roles, collections, or Ansible Vault. This is ansible-core (the engine + `ansible.builtin`), NOT generic configuration-management theory and NOT the bundled community `ansible` package's collection modules.

Run a Socratic teaching loop that quizzes you on a coding session until you have confirmed mastery of every concept. Use when you want to learn or lock in what happened in a Claude Code session, understand a transcript, PR, or design decision, be quizzed/tested/drilled on recent work, or prepare to teach someone else — or when the user runs `/teach` or asks to be taught or quizzed on a topic. Sources sessions from ~/.claude/projects/, tracks a per-concept checklist (Problem / Solution / Broader Context), and never wraps up until every item is confirmed.

A fast, user-friendly command-line tool for finding files and directories by name — a simpler, faster `find` replacement. Use when searching for files or directories by name or regex/glob pattern, filtering results by type, extension, size, or modified time, respecting .gitignore, or running a command per result with -x/-X. Triggers on mentions of the fd command, fdfind, "find files named", "search for a file", or replacing `find`. This is the fd CLI tool, NOT the file-search MCP server.

A fast, gitignore-aware command-line tool for recursively searching file contents by regular expression — a smarter, faster `grep`. Use when searching code or text for a pattern across a directory tree, filtering matches by file type or glob, listing/counting matching files, doing search-and-replace previews, or running gitignore-aware code search. Triggers on mentions of ripgrep, the `rg` command, "search the codebase for", "grep for", "find all files containing", or replacing `grep`/`ack`/`ag`. This is the rg CLI tool, NOT the fuzzy-search MCP server (which wraps rg).

Advanced Git CLI mastery, recovery, and troubleshooting (git 2.54+). Use whenever a git task goes beyond the basics — recovering lost commits/branches/stashes (reflog, fsck), undoing a bad reset/merge/rebase/amend, rewriting history (interactive rebase, --autosquash, --update-refs, filter-repo), resolving merge conflicts or using rerere, working with worktrees, bisect, cherry-pick, revert, or stash, understanding refspecs/remotes/push behavior and --force-with-lease, configuring .gitattributes/.gitignore/hooks, exploring git internals (object model, plumbing like cat-file/rev-parse), or diagnosing confusing errors (detached HEAD, dubious ownership, non-fast-forward, CRLF). Trigger even when the user doesn't name the exact command but is clearly stuck or doing something non-trivial in git. For commit-message wording use conventional-commits; for stacked/dependent branches use git-chain; for auto-folding fixups use git-absorb; for Jujutsu use jj.

Extract and document writing styles from source texts into reusable style guides. Use when the user wants to analyze an author's voice, create a style guide from a book or document, capture writing patterns for replication, or build a writing style rubric. Triggers on 'extract style', 'analyze writing style', 'capture voice', 'style guide from', or 'writing style analysis'.

Write content using stored writing styles from the writing-styles/ collection. Use when the user wants to write in a specific voice, apply a stored style, list available writing styles, or evaluate text against a style rubric. Triggers on 'write in [style] style', 'use [style] voice', 'list writing styles', 'apply style', or 'evaluate my writing'.

Plain-text double-entry accounting with hledger. Use when recording transactions, checking balances, generating financial reports (balance sheet, income statement, cashflow), importing CSV bank statements, budgeting, tracking time, managing multiple currencies, or doing year-end closing. Triggers on mentions of hledger, ledger-cli, journal files, double-entry accounting, balance assertions, or plain-text accounting.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Create and manage Anki flashcards via the AnkiConnect API. Use when the user wants to create flashcards, add cards to Anki, manage Anki decks, review Anki statistics, or interact with Anki in any way. Requires Anki desktop app running with AnkiConnect add-on installed.

Build and refactor CLIs for AI agent compatibility. Use when making command-line interfaces machine-readable, adding structured JSON output, hardening inputs against hallucinations, implementing safety rails like dry-run flags, adding schema introspection, or designing multi-surface architectures (CLI + MCP).

Interact with local Chrome browser session via Chrome DevTools Protocol. Use when asked to inspect, debug, or interact with a page open in Chrome, take screenshots of browser tabs, read accessibility trees, evaluate JavaScript, click elements, navigate pages, or automate browser interactions in a live Chrome session.

Format git commit messages following Conventional Commits 1.0.0 specification. Use when the user asks to commit changes, create a git commit, or mentions committing code. Ensures consistent, semantic commit messages that support automated changelog generation and semantic versioning.

Guide AI-assisted UI generation toward enterprise-grade, intentional design. Use when building user interfaces, creating dashboards, designing enterprise software or SaaS applications, generating frontend code with styling, or when the user asks for design help. Enforces principles inspired by Linear, Notion, Stripe, and Vercel.

Automatically fold uncommitted changes into appropriate commits on a feature branch. Use when applying review feedback, fixing bugs in feature branches, or maintaining atomic commit history without manual interactive rebasing. Particularly useful for making corrections to recent commits without creating messy "fixes" commits.

Manage and rebase chains of dependent Git branches (stacked branches). Use when working with multiple dependent PRs, feature branches that build on each other, or maintaining clean branch hierarchies. Automates the tedious process of rebasing or merging entire branch chains.

Drive Google Workspace from the terminal with the `gog` CLI (gogcli). Use when sending or searching Gmail, managing labels/drafts/filters, creating or updating Google Calendar events (including recurring, focus-time, or out-of-office), uploading/downloading/sharing Google Drive files, reading or writing Google Sheets, editing or exporting Google Docs and Slides, managing Google Contacts or the Workspace directory, working with Google Tasks (including recurrence), posting to Google Chat spaces, running Apps Script functions, creating Forms, administering Workspace users/groups via a DWD service account, or when building agent workflows that need multi-account Google access with JSON output, OAuth/ADC/service-account auth, Pub/Sub watches, email-open tracking, or an `--enable-commands` sandbox. Invoke any time the user mentions gogcli, the `gog` command, a Google CLI, or a one-stop terminal tool for Gmail/Calendar/Drive/Docs/Sheets/Slides/Chat/Tasks/Contacts/Classroom/Forms/Keep/Admin.

Write mathematics in a long-form, understanding-focused style with detailed proofs and rich exposition. Use when explaining mathematical concepts, writing proofs, tutoring math, creating educational math content, or when the user asks for mathematical explanations. Inspired by Jay Cummings' Real Analysis and Chartrand's Mathematical Proofs. Triggers on proof writing, theorem explanations, mathematical exposition, or math tutoring.