{
"markdown": "\n\n# agenticat\n\n**Agents & Skills that I use, compatible with most AI coding tools** \n\n[](LICENSE)\n[](#skills)\n[](#agents)\n\n
\n\n---\n\nEverything here (should be) compatible with **opencode**, **Claude Code**, **Gemini CLI**, **Codex CLI**, **Cursor** and 73+ skill-compatible environments.\n\nSkill and Agent frontmatter is in Claude Code compat format by default; use `npx skills` and/or bundled bun script to make them compatible with other harnesses (will do it in a better way when I decide how). \n\n**[Contributions](CONTRIBUTING.md) are welcome!**\n\n## Install\n\n> [!NOTE]\n> Two plugins ship separately: `agents` and `skills`. Install one or both; on opencode/codex, cherry-pick by name with `./install.sh --to opencode prober web-researcher` (list with `./install.sh --list`). \n\n\nClaude Code
\n\n#### Agents\n\n```\n/plugin marketplace add uwuclxdy/agenticat\n/plugin install agents@agenticat\n```\n\n#### Skills\n\n```\n/plugin install skills@agenticat\n```\n\nor cherry-pick with `npx skills`:\n```\nnpx skills add uwuclxdy/agenticat --skill \n```\n \n\n\nGemini CLI
\n\n#### Agents\n\n```\ngemini extensions install https://github.com/uwuclxdy/agenticat --auto-update\n```\n\n> [!WARNING]\n> Gemini's extension-bundled subagents are a preview feature, verify if it works on your version.\n\n#### Skills\n\n```\nnpx skills add uwuclxdy/agenticat -a gemini\n```\n \n\n\nopencode
\n\n#### Agents\n\n```\ngit clone https://github.com/uwuclxdy/agenticat && cd agenticat\n./install.sh --to opencode\n```\n\n#### Skills\n\n```\nnpx skills add uwuclxdy/agenticat -a opencode\n```\n \n\n\nCodex CLI
\n\n#### Agents\n\n```\ngit clone https://github.com/uwuclxdy/agenticat && cd agenticat\n./install.sh --to codex\n```\n\n#### Skills\n\n```\nnpx skills add uwuclxdy/agenticat -a codex\n```\n \n\n\nCursor
\n\n#### Agents\n\nCursor reads `agents/` as claude format; copy files into `.claude/agents/` or `.cursor/agents/`.\n\n#### Skills\n\n```\nnpx skills add uwuclxdy/agenticat -a cursor\n```\n \n\n\nanything else (73+ harnesses)
\n\nno subagents yet.\n\n#### Skills\n\n```\nnpx skills add uwuclxdy/agenticat\n```\n \n\nOr just download into `~/.claude/skills/`, but then you won't get updates.\n\n## Update\n\n- Claude Code: `/plugin update agents@agenticat` and/or `/plugin update skills@agenticat` (every commit is a new version)\n- Gemini CLI: automatic with `--auto-update`\n- opencode / Codex: `./install.sh --update` pulls and reconverts what you installed\n- Skills: `npx skills update`\n\n## Skills\n\nInstall any skill with the same command, swapping the final name (see [Install](#install) for other harnesses):\n\n```\nnpx skills add uwuclxdy/agenticat --skill \n```\n\nThe `used by` column is a soft link: when an agent's task touches what a skill covers it loads that skill for the deeper method. Cherry-pick either one alone and the loader falls back to its own built-in checklist.\n\n| skill | origin | used by | what it does |\n|---|---|---|---|\n| `clean-code` | `s4.codes` | `python-reviewer`, `ts-reviewer` | language-agnostic readability and naming conventions |\n| `name-check` | original | standalone | proposes project or crate names, checks each for availability on registries and domains (needs `bun`) |\n| `parity-gap` | original | standalone | diffs your project against a reference or spec, writes the gaps as tasks (calls `todo`) |\n| `todo` | original | `parity-gap` | turns findings and loose notes into a clean `docs/todo.md`, each task pickup-cold ready |\n| `askama` | original | `rust-pro` | conventions and reference for the askama Rust templating crate |\n| `maud` | original | `rust-pro` | conventions and reference for the maud Rust `html!` macro |\n| `cargo-toml-optimization` | original | `rust-pro` | tunes `Cargo.toml` and `.cargo` build profiles and features for faster compiles |\n| `clean-rust` | original | `rust-pro` | idiomatic Rust conventions, modular: core rules plus per-domain references for errors, async, concurrency, unsafe, security, observability, testing, perf |\n| `threat-modeling` | original | `threat-modeler` | STRIDE and attack trees, each requirement mapped to a mitigation |\n| `clean-shell` | `wshobson` | `shell-pro`, `shell-reviewer` | defensive live-system idioms, `.shellcheckrc` discipline, bats-core testing; core rules plus a reference for each |\n| `clean-flutter` | original | `flutter-pro`, `flutter-reviewer` | idiomatic Flutter/Dart conventions, modular: core rules plus per-domain references for architecture, state management, navigation, models/serialization, testing, pitfalls, packages |\n| `emulator-testing` | original | `mobile-tester` | drives Android emulators and iOS simulators from the CLI: headless boot, adb/simctl primitives, Flutter test layers, screenshot verification |\n| `skill-routing-audit` | original | standalone | audits a skill for routing gaps: boundary and negative questions that bury the real answer |\n| `mcp-stateless` | original | standalone | the MCP `2026-07-28` stateless revision: what the handshake removal changed, the wire contract, multi round-trip requests, and how to migrate a server or client off session-era code |\n| `docs-sync` | original | `docs-reconciler` | reconciles README, wiki, docs, and CLAUDE.md (plus agent/skill prompt files) with what the code actually does |\n| `ratatui-patterns` | original | `ratatui-pro` | modern ratatui built-ins, TestBackend render tests, a version-upgrade reference pack |\n| `webapp-testing` | `anthropics` | `webapp-tester` | Playwright toolkit for testing local web apps, screenshots and console logs |\n| `ux-patterns` | `synsation_` | `ux-pro`, `ux-reviewer`, `flutter-pro` | how a UI behaves: the five screen states, loading and error patterns, form validation and submission, undo versus confirmation, cognitive-load laws, with WCAG wiring and a mobile reference |\n| `terminal-ux` | original | `ratatui-pro` | how a CLI or TUI behaves: exit codes, stdout vs stderr, TTY detection, `NO_COLOR`, help and destructive-action conventions, panic-safe terminal restore, terminal screen-reader practice |\n\n### origin\n\n- `wshobson`: [wshobson/agents](https://github.com/wshobson/agents) (MIT)\n- `s4.codes`: **content from [@s4.codes](https://www.tiktok.com/@s4.codes) (TikTok) as a Claude Skill**; Gemini transcribes the videos, Opus condenses them into the skill.\n- `synsation_`: **content from [@synsation\\_](https://www.tiktok.com/@synsation_) (TikTok) as a Claude Skill**, the `Build for Good UX` series, parts 1 to 16; same pipeline as `s4.codes`. Its accessibility reference is authored here from W3C sources instead, since the series does not cover accessibility.\n- `original`: self-authored\n- `anthropics`: [anthropics/skills](https://github.com/anthropics/skills) (Apache-2.0), vendored with its LICENSE and a modification note\n- `clean-rust` distills the author's own Rust conventions; a dozen standout rules were re-authored after studying [leonardomso/rust-skills](https://github.com/leonardomso/rust-skills) (MIT)\n\n## External skills\n\nEvery third-party-origin item in the tables above is a copy this repo owns: the `wshobson` skills and `-pro` agents are rewrites, `webapp-testing` is vendored with its license, `clean-code` is distilled from transcripts. Installing them never pulls from the original source.\n\nThe one exception worth a look on its own: the [obra/superpowers](https://github.com/obra/superpowers) skills `systematic-debugging`, `test-driven-development`, and `verification-before-completion` (MIT) are not bundled here and install pristine from source with `npx skills add obra/superpowers --skill `.\n\n## Agents\n\nInstall with the Claude Code plugin (see [Install](#install)) or cherry-pick onto another harness with `./install.sh --to `. Every definition is canonical Claude Code frontmatter; the converter keeps a read-only agent read-only (no Write or Edit) wherever it lands.\n\n| agent | model | loads if installed | what it does |\n|---|---|---|---|\n| `python-reviewer` | inherit | `clean-code` | read-only Python diff/PR review with file:line and severity |\n| `ts-reviewer` | inherit | `clean-code` | same for TypeScript and JavaScript |\n| `docs-extractor` | inherit | none | digests a file or doc set into a brief, keeps raw bytes out of your context; coverage mode proves a doc safe to delete or merge |\n| `spec-propagator` | opus | none | folds a decided spec into a design doc in that doc's own voice |\n| `threat-modeler` | opus | `threat-modeling` | STRIDE and attack-tree threat model, writes one doc |\n| `prober` | haiku | none | runs a build, test or lint, returns pass/fail instead of the full log |\n| `web-researcher` | inherit | none | runs one research question through many searches, returns a cited markdown brief |\n| `webapp-tester` | inherit | `webapp-testing` | drives a local app via Playwright and judges each screenshot by eye; pass/fail with evidence, design-contract checks on request |\n| `shell-pro` + | inherit | `clean-shell` | writes or refactors bash or POSIX sh, verifies with shellcheck |\n| `golang-pro` + | inherit | none | one module-sized Go task, proven with the repo's gate and race detector |\n| `rust-pro` + | inherit | `clean-rust`, `cargo-toml-optimization`, `askama`, `maud` | one Rust task against the repo's cargo and clippy gate |\n| `c-cpp-pro` + | inherit | none | C and C++ with explicit ownership, sanitizers wired to the repo build |\n| `flutter-pro` | inherit | `clean-flutter`, `ux-patterns` | one Flutter/Dart task against the repo's analyze/format/test gate |\n| `flutter-reviewer` | opus | `clean-flutter` | read-only Flutter/Dart diff/PR review with file:line and severity |\n| `mobile-tester` | sonnet | `emulator-testing` | drives a local Android AVD or iOS simulator, reports pass/fail with screenshots |\n| `rust-reviewer` | opus | `clean-rust` | read-only Rust diff/PR review: correctness, safety, async, invariants |\n| `shell-reviewer` | sonnet | `clean-shell` | read-only shell/bash review: quoting, error handling, injection, portability |\n| `root-cause-investigator` | opus | none | evidence-first root-cause hunt for disputed regressions, returns a hypothesis ledger |\n| `tui-tester` | sonnet | none | drives a TUI/CLI in a real tmux pty like a user, reports pass/fail with captured screens |\n| `frontend-builder` | sonnet | none | builds self-contained single-file frontends: canvas/WebAudio demos, SVG and favicon glyphs |\n| `docs-reconciler` | sonnet | `docs-sync` | reconciles README, docs, CLAUDE.md, and agent/skill prompt files with the code, edits docs only |\n| `ratatui-pro` | sonnet | `ratatui-patterns`, `terminal-ux` | one Rust TUI task with current ratatui built-ins plus TestBackend tests |\n| `ux-pro` | sonnet | `ux-patterns` | builds a screen's loading, success, error, empty and partial states with the repo's own components |\n| `ux-reviewer` | opus | `ux-patterns` | read-only audit of what a UI does: missing states, silent failures, error copy, validation gaps |\n\n- `loads if installed` is soft, has the same fallback rule as the skills table. \n- `+` began as [wshobson/agents](https://github.com/wshobson/agents) personas (MIT), rewritten and debloated.\n\n## External repos checked\n\nRepos evaluated as skill or agent sources. Adopted ones are marked; the rest were passed over for the reason given.\n\n\nshow the list
\n\n- `wshobson/agents`: adopted. source of the shell skills and the `-pro` agents, all rewritten and debloated.\n- `obra/superpowers`: linked external, see [External skills](#external-skills).\n- `anthropics/skills`: `webapp-testing` vendored (Apache-2.0).\n- `vercel-labs/skills`: adopted as the `npx skills` installer, not a content source.\n- `VoltAgent/awesome-claude-code-subagents`: 100+ subagent index, generic, nothing worth vendoring.\n- `0xfurai/claude-code-subagents`: `playwright-expert` too thin, wrote own `webapp-tester`.\n- `vijaythecoder/awesome-claude-agents`: `frontend-developer` compared, no fit.\n- `yusuftayman/playwright-cli-agents`: CLI-focused, replaced by own `webapp-tester`.\n- `rshah515/claude-code-subagents`: browsed, generic bulk.\n- `contains-studio/agents`: `frontend-developer` compared, redundant.\n- `lodetomasi/agents-claude-code`: `playwright-pro` fetched, redundant.\n- `darcyegb/ClaudeCodeAgents`: `qa-engineer` link 404'd, dead.\n- `leonardomso/rust-skills`: vendored once then delisted. one fat `SKILL.md`, not modular enough. its standout rules were later re-authored into the first-party `clean-rust`.\n- `actionbook/rust-skills`: competing rust system, passed over.\n- `vercel-labs/open-plugin-spec`: spec only, nothing implements it.\n- `affaan-m/ecc`: \"67 agents, 271 skills\" claims unverified.\n- `numman-ali/openskills`: skills-only, no agent definitions.\n- `anthropics/claude-code`: read for plugin-marketplace mechanics (issue #20301), not content.\n- `google-gemini/gemini-cli`: read for extension and subagent mechanics.\n- `charmbracelet/crush`: subagent format unshipped at research time.\n- `cursor/plugins`: spec has no agent-definition concept.\n- `mohitagw15856/pm-claude-skills`: `--agent` picks a destination harness, not agent defs.\n- `dmicheneau/opencode-template-agent`: template, not an installer.\n- `antfu/skills-npm`: competing install-from-npm concept.\n- `VintLin/skill-flow`: competing cross-agent skill manager.\n- `hesreallyhim/awesome-claude-code`: index only, used for discovery.\n- `sickn33/antigravity-awesome-skills`: bulk dump, no curation.\n- `google-labs-code/stitch-skills`: unrelated, cited only for an issue.\n- `gemini-cli-extensions/security`: gemini-specific.\n- `Piebald-AI/claude-code-system-prompts`: system-prompt mirror, reference only.\n- `agentskills/agentskills`: spec site, not content.\n\nAbout 30 more were surveyed in a separate trending/growth analysis, not as adoption candidates, and are omitted here.\n\n \n\n## License\n\nMIT. \n\nSkills with a third-party origin keep their upstream licenses (see [origin](#origin)).\n",
"bytes": 14227,
"sha": "94fc0e78f24b3e338012e2b08924a86cff86958d57c1853f30a79d8cea9f0996",
"repo_slug": "uwuclxdy/agenticat",
"fonte": "repo",
"truncated": false,
"api": "https://agentalog.com/api/listings/plg_uwuclxdy_agenticat_12015b84/readme"
}