ai-skills

Plugin ai-skills from marketplace bgelis-ai-skills in BenjaminG/ai-skills. Commands read from the repo's manifest — yesterday.

Skills (60)

• acli

  Use Atlassian CLI (acli) to manage Jira work items, projects, and workflows from the command line. Ideal for bulk operations, automation, scripting, and when users request CLI-based Jira interactions. Trigger on requests like "use Jira CLI", "create Jira issues via CLI", "bulk update Jira tickets",

• applying-solid-principles

  This skill should be used when implementing code that requires SOLID principles and clean code practices. It provides detailed guidance on Single Responsibility, Open/Closed, Liskov Substitution, Interface Segregation, and Dependency Inversion principles with comparison examples in TypeScript.

• autobrowse-cc

  Build or harden a reliable browser-automation skill through a self-improving loop, with no Anthropic API key. Each iteration runs a headless `claude -p` inner agent that drives a local isolated Chrome via chrome-cdp, reads the run trace, adds one heuristic to strategy.md, and repeats until the task

• build-in-public

  This skill should be used when the user wants to write their end-of-day / evening "build-in-public" update for a given Linear project — 2-3 impact-focused bullets plus a Next line synthesized from the project's Linear activity and the repo's git/PR shipping evidence, ready to paste into the pod thre

• chrome-cdp

  Interact with a local Chrome browser session via Chrome DevTools Protocol (only on explicit user approval after being asked to inspect, debug, or interact with a page open in Chrome). Pass `--isolated` to drive a sandboxed Chrome with a dedicated profile instead of the user's real browser.

• ci-watch

  Watch a PR's CI hands-free after a push. A persistent Monitor streams each check as it lands; the skill auto-handles the mechanical (rebase a clean conflict, fix lint) and STOPs on any judgment (failing tests, a stack restack, going ready). Triggers on "watch CI", "surveille la CI", "attends la CI",

• confluence-cli

  Use confluence-cli (NPM package) to manage Confluence content, pages, and spaces from the command line. Ideal for documentation workflows, bulk content operations, page migration, and when users request CLI-based Confluence interactions. Trigger on requests like "use Confluence CLI", "create Conflue

• consensus

  Run a structured prompt N times and consolidate via deterministic vote (self-consistency, Wang et al. 2022). Use when output stability matters more than latency — code review, finding extraction, classification, triage. Triggers on "self-consistency", "consensus", "vote on", "stable output", "run N

• create-verification-skill

  Generates a project-local Claude Code skill (.claude/skills/verify-<app>/) that drives the repo's app the way a user does and captures proof, for any language, framework, or platform. This skill should be used for /create-verification-skill, \"make a verify skill for this repo\", or when a project h

• hooks

  Create and manage Claude Code hooks for automating workflows, validating inputs, and customizing behavior. Use when users request to create hooks, set up automation for tool usage (PreToolUse, PostToolUse), add session context (SessionStart), validate prompts (UserPromptSubmit), control stopping beh

• maintain-verification-skill

  Periodic pass that keeps a project's verification skill and feature map honest: parallel Explore subagents read the source per feature, one live session drives every feature, at most one PR of proven corrections. This skill should be used for /maintain-verification-skill or \"audit the verify skill\

• mimic

  Extract and reproduce a specific person's coding style from a git repository. Use this skill whenever the user wants to analyze how a teammate codes, build a reusable style profile from someone's commits or PRs, make code generation match a particular author's conventions, onboard into a codebase by

• qa-run

  Execute a qa-plan interactively, autonomously driving Chrome via chrome-cdp by default. Each run is posted as its own sub-issue/file under the plan (--tracker linear|github|local), re-verifies every scenario, and surfaces prior runs' open findings. A persisted playbook (site knowledge + per-scenario

• quality-gate

  Run parallel quality reviews (React, SOLID, Security, Simplification, Slop) on branch changes and auto-fix issues

• rca

  Investigate a High/Urgent bug on Linear and post a structured root-cause comment in the team's format (Root cause → Timeline of PRs → Category → what the fix does → follow-up). Drafts in chat first; posts to Linear only on explicit approval. This skill should be used when asked to document why a bug

• refactor-instructions

  Audit and refactor CLAUDE.md, AGENTS.md, and .claude/rules/ files to follow progressive-disclosure principles. This skill should be used when the user wants to reorganize, clean up, or optimize their instruction files and rules by splitting monolithic files into a minimal root, domain-scoped rules w

• second-pass

  This skill should be used when the user wants to rebuild a just-completed implementation from scratch on a fresh branch, applying everything learned during the first pass. The first pass is exploration; the second pass is the real solution. Triggers on "second pass", "rewrite now that you know", "re

• skill-creator

  Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

• swiftui-performance

  Diagnose and fix SwiftUI performance issues including hangs, hitches, slow launches, excessive view redraws, and app terminations. This skill should be used when profiling SwiftUI apps, investigating UI responsiveness problems, optimizing render performance, reducing launch time, or fixing memory-re

• ubiquitous-language

  Extract a DDD-style ubiquitous language glossary from the current conversation, flagging ambiguities and proposing canonical terms. Saves to UBIQUITOUS_LANGUAGE.md. Use when user wants to define domain terms, build a glossary, harden terminology, create a ubiquitous language, or mentions "domain mod

• ui-skills

  Opinionated constraints for building better interfaces with agents.

• babysit-prs

  Drive every open PR to merge-ready without polling by hand. Follows the transitions the shared pr-dash Scan service writes, through a persistent Monitor, and hands each PR that has bot feedback or a red check to its own subagent, which answers, folds and pushes on its own. Human reviews are counted,

• backend-developer

  Specialist for TypeScript backend development including Node.js servers, NestJS applications, APIs, databases (PostgreSQL, MongoDB, Redis), authentication, testing, deployment, server-side code, microservices, database queries, migrations, controllers, services, repositories, and middleware. Use for

• code-slop

  This skill should be used when the user wants to find AI-generated code slop — unnecessary comments, defensive checks, `any` casts, and style inconsistencies — in a branch's changes. Produces suggestions only; does not apply edits.

• codex-cli

  Use OpenAI Codex CLI in non-interactive mode for automated code analysis, review, and programmatic task execution. Trigger on requests like "use Codex to analyze", "run codex exec", "codex code review", or when users want AI-powered code analysis without interactive prompts. Ideal for automation wor

• commit

  Stage and commit pending changes with an automatically generated concise message. This skill should be used when the user asks to "commit", "stage and commit", "make a commit", or "save changes to git". Does not push or create PRs.

• daily-update

  Draft Benjamin's daily Slack update for the Naboo team from Linear. Pulls his issues closed since the previous work day (DONE), currently started (IN PROGRESS), and optionally blocked (BLOCKERS), formats them in the team's Slack style, and iterates in chat until approved. Triggers on "daily update",

• dream

  This skill should be used to run a reflective consolidation pass over persistent memory files — orienting on what already exists, gathering recent signal from session logs and transcripts, merging new facts into topic files rather than creating near-duplicates, and pruning the index of stale or cont

• elevate

  This skill should be used when the user wants the highest-leverage technical or architectural changes for their project, ranked from what breaks in production (recurring bugs, incidents, slow paths, frictions reported in Slack, Linear or Datadog) and from the code itself, ending on one recommended c

• explain-plainly

  Re-explain a technical concept, code path, or review finding you just produced in plain language, for a reader who is not in that domain. Use when the user says it's "charabia"/gibberish, asks you to "explique simplement", "traduis en clair", "je comprends pas", "explain like I'm not a specialist",

• fixup

  Fold an uncommitted fix into the commit that introduced the code, so the branch reads as if the bug never existed. Finds the introducing commit, runs `--fixup` plus `--autosquash` (or `--amend` at the tip), then force-pushes with lease. Use for every fix on an unmerged branch, whether it came from r

• frontend-developer

  Specialist for modern React applications, TypeScript frontend code, UI development, TanStack Query, component creation, CSS styling, responsive design, accessibility, form validation, state management, routing, and performance optimization. Use for ANY work involving .tsx/.jsx files, React component

• gate-wf

  Workflow-native quality gate for branch changes — parallel reviewers (Bug, SOLID, Security, Simplify+Slop, optionally React/a11y/i18n/migration) with tier-scaled adversarial verify, CLAUDE.md/ADR enforcement, and a stable PASS / PASS WITH NOTES / FAIL verdict. Read-only. Runs from a static shipped w

• gate

  Harness-portable quality gate for branch changes — spawns parallel reviewer subagents (Bug, SOLID, Security, Simplify+Slop, optionally React/a11y/i18n/migration) with tier-scaled adversarial verify, CLAUDE.md/ADR enforcement, and a stable PASS / PASS WITH NOTES / FAIL verdict. Read-only. Runs on any

• harness-review

  Turn one PR's review comments into harness fixes — classify each cause by the most solid place it can live (codebase, lint, rules, skill), then draft the fixes in a separate PR.

• herdr-mod

  Builds a mod for the Herdr desktop app (a panel, sidebar, popup or key-driven view in ~/.config/herdr-app/mods/) and, with --publish, shares it on herdr.dev/mods through a public GitHub repo. This skill should be used when asked to make, change or publish a Herdr mod or panel.

• humanizer

  Remove signs of AI-generated writing from text and progressively learn the user's personal writing style. Use when editing or reviewing text to make it sound natural and human-written, or when the user says "humanizer init" / "humanizer update: ..." to customize the skill. Based on Wikipedia's "Sign

• innovate

  This skill should be used to rank evidence-backed product additions or capability changes and recommend one winner, across either the full product or an optional focus such as activation, retention, differentiation, AI, onboarding, or power users. Use `elevate` for architecture, tooling, or DX.

• interview

  Interview user to clarify any topic - exploring codebase, investigating issues, planning features, understanding requirements, or drilling into plans. Socratic questioning to uncover details.

• investigate

  Pull the full context of a Linear/Jira issue or Slack thread — ticket, thread, prior work, code — then plan next steps.

• jev-verify

  This skill should be used to measure whether TypeSafe's Jev model can replace gate-wf's sonnet skeptics in the Verify phase. Replays findings from gate-wf state files against Jev — one request per finding, four typed questions — and reports where the two disagree, plus cost and latency for both path

• orchestrate

  Run a dependency-ordered list of Linear tickets through Herdr (herdr-app with hc, or the older herdr multiplexer). One tab, one worktree, one agent per layer, committed and checked before the next layer starts, PRs opened at the end. Use when asked to orchestrate, run, or dérouler a stack of tickets

• pr-challenge

  This skill should be used when reviewing someone else's pull request the way a colleague would — a few questions that challenge the approach, the reuse, the shape, the intent or the naming — or an LGTM when the PR reads clean. Drafts the comments and hands them to pr-comment to post. Triggers on "re

• pr-comment

  This skill should be used when reviewing someone else's pull request and you want drafted review comments posted to GitHub — either gate / gate-wf findings (tier-marked) or pr-challenge's reviewer questions (unmarked) — each drafted through humanizer and posted as a standalone inline comment at its

• pr-create

  Publish a pull request with automated type detection, Linear/Jira linking, PR templates, and Slack review message.

• pr-demo

  Records a short demo video of a pull request's feature in the local app through Playwright MCP, then attaches it to the PR description. This skill should be used when asked for a demo video of a PR or feature ("vidéo de démo", "parcours le flow… donne-moi la vidéo"), to re-record or update the video

• pr-explain

  Open a herdr tab where a fresh Claude session triages a PR and explains its held thread in plain words.

• pr-feedback

  Triage a pull request — sort every unresolved review item and failing check into P1 / P2 / Nit with a disposition, for `pr-respond` to act on. Use when triaging a PR, asking what is blocking a PR, or given a bare PR number or URL.

• pr-respond

  This skill should be used after pr-feedback triage, when the user has picked which review items to act on and wants them applied and answered — apply the agreed code change per item, then post the reply, reaction, and (for inline threads) thread resolution in one confirmed batch. Triggers on "apply

• pr-risk-score

  Score one Naboo pull request and render a risk report (score, auto-approval eligibility, confidence, findings), posted as a single PR comment only after explicit confirmation.

• pr-shots

  Takes screenshots of a pull request's change in the local app through Playwright MCP, as before/after pairs or as a gallery of a feature's states, then attaches them to the PR description. This skill should be used when asked for screenshots or captures of a PR ("avant/après", "before/after", "scree

• pr-try

  Hands the local app, running a PR's or branch's code, to the user for a manual test, one scenario at a time from its start state, in a visible browser. This skill should be used when the user wants to test a PR, a branch or a fix by hand ("je veux tester", "tester à la main").

• product-qa

  Announce that a feature is ready for manual QA on its ephemeral environment by creating a "[PRODUCT QA]" sub-issue on Linear with the QA label applied at creation, so the team's product_bot auto-posts the ephemeral-env links to the #product-qa Slack channel. Drafts the body from the PR diff and the

• qa-plan

  Draft a manual QA test plan for a pull request, tailored for a PM or QA tester (not a developer). Auto-detects the current PR from the branch, or takes a PR number/URL. Produces steps that can be run with only a browser and the app's admin UI — no terminal, SQL, GraphQL, or code access. Always rende

• retrospective

  Reflect on the work done on the current branch vs main — mine devsql history, git log, and session transcripts for back-and-forths, rollbacks, and wasted time, then interview the user to distill durable lessons into a local-ignored CLAUDE.local.md.

• review-mining

  Mine human review comments from merged PRs into recurring themes and propose where each rule should live (lefthook/Danger, CLAUDE.md, gate-wf reviewer). Report only, no edits.

• setup-ai-skills

  Configures a host repo for the ai-skills that read it, by writing and proving its `docs/agents/` docs - `local-app.md` and its scripts (pr-demo, pr-shots, pr-try, showcase) and `signals.md`, where production symptoms live (elevate). This skill should be used for /setup-ai-skills, when one of those s

• showcase

  Produces a narrated, branded demo video of a feature, a pull request, a branch or an area of an app, for the team (a pod's weekly update, a feature launch), from the local app filmed through Playwright MCP. The script is approved before anything is filmed; the voice-over comes from the ElevenLabs MC

• skill-dream

  Improve one ai-skills skill from its own past runs — cut every run out of the local Claude Code transcripts, find where the user had to correct it, chain it by hand, or watch a prescribed command fail, drop what later versions already fixed, and open a draft PR with the recurring fixes. This skill s

• triage-findings

  This skill should be used to act on the findings of a previous gate-wf run — select a subset by tier, by rule family, by id, by file, or by a free-text intent ("only the ones that shrink the code"), then fix them after one confirmation. Triggers on "/triage-findings", "fix the blockers", "clean up t

Subagents (13)

• a11y-reviewer

  Reviews JSX/TSX in a code diff for accessibility regressions. Invoked by the gate / gate-wf review skills when .tsx/.jsx files are present.

• bug-reviewer

  Reviews a code diff for logic bugs, spec-code mismatches, and cross-file behavioral parity gaps. Invoked by the gate / gate-wf review skills.

• context-checker

  Checks reviewer findings against project context (CLAUDE.md, ADRs, Linear, PR comments, past sessions) and synthesizes findings for documented-rule violations. Invoked by the gate / gate-wf review skills.

• i18n-reviewer

  Reviews JSX/TSX in a code diff for i18n/l10n regressions. Invoked by the gate / gate-wf review skills when .tsx/.jsx files plus an i18n library are present.

• jev-runner

  Runs jev_verify.py batch over prepared finding files and returns its JSON verbatim. Never edits findings, never invents scores. Invoked by gate-wf's verify phase when --jev is active.

• migration-reviewer

  Reviews database migrations and bulk-update operations in a code diff for data-loss and rollback risks. Invoked by the gate / gate-wf review skills when migration files or bulk-update APIs are present.

• ponytail-reviewer

  Reviews a code diff for over-engineering — reinvented stdlib, code the repo already has, unneeded dependencies, speculative abstractions, dead flexibility. Invoked by the gate / gate-wf review skills.

• react-reviewer

  Reviews a code diff for React/Next.js best-practice violations and component-composition smells. Invoked by the gate / gate-wf review skills.

• security-reviewer

  Reviews a code diff for security vulnerabilities and emits structured findings. Invoked by the gate / gate-wf review skills.

• simplify-reviewer

  Reviews a code diff for over-engineering, dead code, and simplification opportunities. Invoked by the gate / gate-wf review skills.

• skeptic

  Adversarially refutes a single reviewer finding. Returns refuted=true by default when uncertain. Invoked by the gate / gate-wf review skills during the verify phase.

• slop-reviewer

  Reviews a code diff for AI-slop patterns — defensive checks for impossible states, comment noise, `any` casts, style drift. Invoked by the gate / gate-wf review skills.

• solid-reviewer

  Reviews a code diff for SOLID violations and emits structured findings. Invoked by the gate / gate-wf review skills.

Similar plugins