Menu
Skills dans ce dépôt
SkillsMP a collecté 447 skills depuis jleechanorg/claude-commands. Ouvrez un skill pour examiner sa source et ses détails.
jleechanorg/claude-commandsAffichage de 40 skills collectés sur 447.
Verify that a claimed frontend-fix (CSS, JS behavior, layout, scroll, click handler, asset, computed style) is actually live on a deployed URL. Produces a yes/no verdict with three layers of proof — (1) merged code on origin/main, (2) fix-markers present in…
Langue du texte source : anglais
Capture screenshots of web pages for OG thumbnails, social previews, evidence, QA, or review. Covers Playwright CLI (primary), browser tools (fallback), local serving, meta tag injection, and gitignore PNG exceptions. v1.3.0 (2026-07-21) adds:…
Langue du texte source : anglais
Write implementation plans: bite-sized tasks, paths, code.
Langue du texte source : anglais
Diagnose why a Spend Alert (GitHub Actions MTD/7-day breach, MCP Agent Mail relay,
Langue du texte source : anglais
When asked to verify whether a Claude Code feature works (especially slash commands, dialogs, pickers, status indicators), spawn a real interactive TUI session in cmux — never use `claude --print "/feature"` as a test, because --print is non-interactive and…
Langue du texte source : anglais
Use when the user invokes /up or asks to persist a user-global coding-agent rule or preference across Claude, Codex, Gemini, Cursor, or Hermes.
Langue du texte source : anglais
Pre-adoption evaluation recipe for any third-party ML artifact (router, classifier, embedding model, RAG pipeline). Before recommending adoption — or even running it on real production traffic — load this skill and run the 6-gate checklist: (1)…
Langue du texte source : anglais
Verify an automated alert (BQ coverage watcher, MCP Agent Mail, GH Actions billing, infra watcher) or in-house scheduled cron (Daily Bug Hunt, doc-audit, stability-report) before acking it. Cross-checks the alert's named mechanism against live state and the…
Langue du texte source : anglais
When /green on a your-project.com PR fails, the right fix is PR-shape cleanup, not gate adjustment. Use when a PR title claims small scope but the diff is large, or when Green Gate Precheck fails with stale `origin/main`/re-replays-of-merged-PRs pattern.…
Langue du texte source : anglais
Recurring your-project.com workflow for editing `$PROJECT_ROOT/` settings surface (dropdowns, models, BYOK, provider selection) — edit code, run local server, capture browser+API proof. Trigger when the user says 'add a model to settings', 'change a settings…
Langue du texte source : anglais
Read Your Project production Firestore and produce a real-user engagement / retention report for a date window. Use when asked to review last week's real-user activity, analyze WA retention, look at WA prod data, find recent WA campaigns from real users,…
Langue du texte source : anglais
Edit $PROJECT_ROOT/prompts/*.md to fix LLM-spawned behavior surfaced by a campaign — minimal-diff PR against origin/main, no lore dumps, no content deletion. Use when the user reports "the AI keeps doing X" / "spawning Y" / "using Z archetype" / "ignoring…
Langue du texte source : anglais
Safety patterns for working with a populated LLM-wiki / personal knowledge vault (~/llm_wiki, ~/wiki, etc.) — preventing concept-page clobber, enforcing append-not-overwrite, the default-doing-now save location for research artifacts, AND the…
Langue du texte source : anglais
Never just make local edits and stop. Always create a GH issue + bead, dispatch via ao spawn for a PR, and skillify the pattern. Local exploration is fine; local edits without a PR are a process violation.
Langue du texte source : anglais
When a user (often via Slack, GitHub, or email) sends a pre-built patch + a structured instructions file (UPSTREAM-PROMPT.md style) and asks you to open the PR — clone the target repo, branch from origin/main, git am the patch, run the user's listed test…
Langue du texte source : anglais
v2.5.13 overlay — Design Doc Grep Gate (Gate 0) + Tenets/Design-Decision section + empty-commit re-trigger trick (2026-07-28, PRs
Langue du texte source : anglais
Use when multi-model reviewers are auth-walled.
Langue du texte source : anglais
Launch, verify, and regression-protect real Claude Code teammates running in tmux with the `--teammate-mode tmux` primitive. Use when the user wants a real parallel Sonnet Claude Code team (not pseudo-agent narration), when fixing slash commands that drifted…
Langue du texte source : anglais
Sanitize a working personal file (CLAUDE.md, AGENTS.md, skill, README, dotfile) into a publishable public example and ship it to a public repo via a clean PR — replacing every personal/internal reference with clearly-marked placeholders, embedding a reusable…
Langue du texte source : anglais
Design a tabletop campaign bible (D&D 5e, Pathfinder, custom system) for a specific character + setting by running the superpowers brainstorming protocol — multiple-choice questions one at a time, hard-gate no-implementation-before-design-approval — and write…
Langue du texte source : anglais
Analyze existing Your Project campaigns in place to diagnose prompt, agent, or quality issues — without copying or exporting. Per-scene content classification with agent attribution via debug_info.agent_name. Use when the user asks 'why is the LLM doing X in…
Langue du texte source : anglais
Find a specific Your Project campaign by premise — the trope/story the user remembers ("female resurrected OP person", "demon lord reincarnation", "where I was a swords bard in Thay", etc.). Searches BOTH Firestore (live data, paraphrased descriptions) AND…
Langue du texte source : anglais
Diagnose Cloud Run deployment failures for Your Project (and any python-on-Cloud-Run service). Also load for RUNTIME failures when the deployed revision is serving and `/health` returns 200 but the user-facing feature is broken — the proxy may be masquerading…
Langue du texte source : anglais
Diagnose daily cron FAILs for Your Project (Dice Audit, Level Up Test, and any future GCP cron job in the `wa-*-daily-*` family). Triggers when a `[GCP Cron] ... - FAIL (...-YYYY-MM-DD)` email arrives, when the…
Langue du texte source : anglais
Diagnose and fix your-project.com LLM model-selection bugs — wrong model in allowlist, missing from `MODELS_WITH_CODE_EXECUTION`, infinite-loop on code-execution calls, default-model changes that broke production. Triggers when the user reports 'dice stopped…
Langue du texte source : anglais
Consolidate all artifacts for a your-project.com campaign module — the world_reference/ markdown files (V1 campaign bible + V1/V2 god-mechanics specs + Faerûn pantheon docs), its live GitHub PR, the Gemini share-link source, and any $PROJECT_ROOT/ code…
Langue du texte source : anglais
Redesign Your Project campaign-tier mechanics — recurring work on the god/multiverse/divine/sovereign/level-up systems in $GITHUB_REPOSITORY. Triggers: 'redesign X tier', 'rework god mode', 'campaigns less challenging', 'disable tier X', 'no campaign…
Langue du texte source : anglais
Write an outcome-driven goal spec that drives agents toward verifiable outcomes — solid evidence, green-CI PRs, high code quality. Pulls context from the last 30 days of coding-CLI convos (/history) and cross-session memory (/ms) so the goal is grounded in…
Langue du texte source : anglais
Read and steer another cmux terminal tab through the Unix socket.
Langue du texte source : anglais
Audit automation job runs, logs, PR processing metrics, and AI-authored remote commits.
Langue du texte source : anglais
Capture browser traffic with Playwright and generate Python client + MCP tool artifacts from HAR sessions.
Langue du texte source : anglais
API skill for mvp-site-app-dev-i6xf2p72ka-uc.a.run.app. Use when: interacting with this site's APIs, automating workflows, debugging requests. Auth: Unknown — capture may reveal auth mechanism.
Langue du texte source : anglais
General read-only execution tracer. Profiles automate known systems; without a profile the LLM must fresh-investigate entry→hops→blocker from live evidence. Triggers on /callpath, trace execution, where stuck.
Langue du texte source : anglais
Steer work on $USER's MacBook via SSH. Use when asked to run commands, install software, check services, or manage files on the MacBook from jeff-ubuntu or another machine. Auto-detects whether running locally on MacBook (no SSH) or remotely (SSH into MacBook…
Langue du texte source : anglais
Build a developer reproduction bundle for agent-review failures, missed-plan retrospectives, Claude/Codex transcript comparisons, or cases where Anthropic/OpenAI engineers need raw conversation JSONL, subagent state, repo evidence, hashes, and a concise…
Langue du texte source : anglais
Use when writing any spec, design doc, or implementation plan for a non-trivial feature — enforces the three-doc rule (no-code spec, design doc with interfaces, TDD impl plan) with adversarial review gates between stages.
Langue du texte source : anglais
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
Langue du texte source : anglais
Self-check your own completed change before independent review — the pre-review sanity pass. Use when you want to check your work, run checks, validate changes, make sure a change is ready, test it end-to-end, run repo guardrails (lint, typecheck, tests,…
Langue du texte source : anglais
Extract action items from today's Granola/Grain meetings, create Todoist tasks, complete fulfilled tasks, and draft meeting-triggered follow-up emails. Use when running the daily action items cron, post-meeting cron, or when user asks to process meeting…
Langue du texte source : anglais
Private-org (Agnt-F) dispatch plumbing for Agent-Orchestrator. Load this companion to the `agento` skill when dispatching AO work against repos under the user's private `$USER-af` GitHub account / Agnt-F org, which the default AO bot token cannot see.…
Langue du texte source : anglais