メニュー
このリポジトリの skills
SkillsMP は jleechanorg/claude-commands から 447 件の skill を収集しています。skill を開くとソースと詳細を確認できます。
jleechanorg/claude-commands収集済み skill 447 件中 40 件を表示しています。
Run bidirectional alignment between Hermes and cmux-managed coding terminals through a visible Slack thread loop. Use only when the user explicitly requests this steering mode (e.g. steer ao-primary in Slack, bidi cmux, operator prompts). Do not use for…
原文の言語: 英語
Enforce headless browser automation by default for Hermes — Playwright MCP and superpowers-chrome. Use when opening browsers, scraping, UI tests, localhost verification, or any chrome_use_browser / Playwright call. Never headed unless Jeffrey explicitly…
原文の言語: 英語
Use the browserclaw CLI to capture browser traffic, infer API endpoints, generate Python clients, OR decrypt + inject Chrome cookies for session reuse. Use when asked to reverse-engineer a website, generate an API client from a HAR, or reuse a logged-in…
原文の言語: 英語
Publication-ready solo campaign bible from canonical templates (Campaign 9-section bible + Sub-Templates A/B/C; Personality 6-tier dossier + d20). v1.2.0: NPCs evolve via Want/Fear/Boundary state; AI mysteries (3 clues, 3 suspects) target each NPC; plot arcs…
原文の言語: 英語
Design a new version (vN+1) of a recurring user campaign by re-reading every prior version in Google Docs and the wiki, extracting the user's 3 recurring taste-pillars, picking ONE pivot variable per version, writing the new campaign bible to Google Doc AND…
原文の言語: 英語
Delegate coding work to Claude Code CLI routed through the user's bashrc `claudem` (alias `claudeminimax`) wrapper that points at a non-Anthropic provider (typically MiniMax M3). Thin wrapper over the bundled `claude-code` skill — same two modes (print +…
原文の言語: 英語
Durable prevention against silent cloud.superpowers.build SSH-key / enrollment-key / known_hosts expiry — daily launchd watchdog + pre-dispatch probe + autoheal. Pair with umbrella skill `superpowers-cloud-build`.
原文の言語: 英語
Run or maintain the cmux approval worker that scans cmux terminal surfaces for approval dialogs, classifies them with `codex exec`, and sends the matching approval key. Use when testing or operating the launchd-based auto-approver, debugging missed prompts,…
原文の言語: 英語
Find the cmux workspace hosting a given topic (e.g. "agy", "bq", "lvl"), AND any meta-question about cmux workspace finding/steering confusion. Trigger phrases: "find the cmux workspace for X", "which cmux tab is doing X", "find the [topic] workspace",…
原文の言語: 英語
Verified list of community MCP servers for cmux + decision guide on which to install. Trigger when the user asks 'does cmux have an MCP server', 'cmux MCP', 'install cmux MCP', 'use cmux via MCP', 'cmux MCP tools', 'use a real CLI/MCP for cmux instead of the…
原文の言語: 英語
Control cmux terminal multiplexer via its Unix socket API. Use when needing to: (1) List, create, select, or close workspaces; (2) Split panes and manage surfaces; (3) Send text or key presses to terminals; (4) Read terminal screen text from a surface; (5)…
原文の言語: 英語
Daily 08:00 PT cmux terminal surface inventory + classification. Lists every selected cmux workspace/surface, reads each pane briefly, classifies as Healthy/Risky/Blocked, and posts a per-surface digest (with pinned-first priority, full PR hyperlinks, and…
原文の言語: 英語
Install and operate the codex `path-deletion-guard.py` PreToolUse hook that allows deletions inside /tmp (and $TMPDIR) but blocks rm/rmdir/shred/find -delete/git destructive ops/apply_patch deletes/Python os.remove+shutil.rmtree outside those dirs. Emits the…
原文の言語: 英語
Generate images, video, and audio via diffusion workflows.
原文の言語: 英語
Humanize text: strip AI-isms and add real voice.
原文の言語: 英語
54 real design systems (Stripe, Linear, Vercel) as HTML/CSS.
原文の言語: 英語
Ship a CLI lifecycle hook (Stop, PreToolUse, PostToolUse, SessionStart, UserPromptSubmit) that runs correctly across Claude Code, Codex, Cursor Agent, Antigravity (Gemini CLI's antigravity branding), and agy. The four CLIs disagree on (a) where they discover…
原文の言語: 英語
Run or fix the exec-assistant daily cron. Covers resolver-recovery (skill_view ambiguity → read_file canonical path), token resolution, channel-routing with cron-override, decay-verify of yesterday's queue (P96), and GitHub-API rate-limit fallback.
原文の言語: 英語
Prepare {{OWNER_NAME}}'s task list for the day using `~/.hermes/workspace/tasks/current.md` plus his calendars. Use when a cron or direct request asks to prepare today's tasks before {{OWNER_NAME}} starts work; when recurring weekday tasks, due-today backlog…
原文の言語: 英語
Wire any long-running auto-factory daemon (dark-factory /af, or any per-repo factory-style crontab) to post lifecycle events to a Slack channel. Covers: 1) creating the notification channel when one doesn't exist, 2) the four event classes every factory needs…
原文の言語: 英語
Add a new third-party API provider (DeepSeek / GLM / OpenRouter / Minimax / etc.) as a Claude Code and Codex CLI shell wrapper, mirroring the existing `claudem` (Minimax) / `claudeds` (DeepSeek) / `claudeg` (GLM) pattern in `~/.bashrc`. The canonical recipe…
原文の言語: 英語
Daily request/token/cost breakdown across every AI CLI installed on this Mac (Claude Code, Codex, Hermes, OpenCode, Gemini, Kimi, Qwen, Copilot, Goose, Kilo, Droid, OpenClaw, Amp, Codebuff, Pi). Trigger when the user says "how many requests do I do daily",…
原文の言語: 英語
Drive a forked bug-hunt PR (scripts/bug-hunt-daily.sh) to durable green by chunking hermes -z workers under one global deadline, retrying failed chunks on a proven surviving model, repairing truncated JSON, and verifying with real-agent dry runs before…
原文の言語: 英語
Evaluate a third-party AI/ML repo, vendor claim, or article headline — clone, run inference, contract-verify the README vs source vs shipped checkpoint, and publish a self-contained reproducible eval to jleechanorg/evals. Trigger when a user message pairs a…
原文の言語: 英語
Diagnose GCP daily cost spikes from the BQ billing export when the user asks "why is X service so expensive", "why did GCP cost go up", "drill into Cloud Run spend", or any per-service or per-SKU cost breakdown for a GCP project. Encodes the working pattern…
原文の言語: 英語
When gh pr view / gh pr checks / gh run view / gh api returns GraphQL rate-limited, 401 Bad credentials, or 503 No-server-available, this is the diagnostic-and-recovery playbook. Encodes the resolution matrix that distinguishes the three failure modes by…
原文の言語: 英語
Operate the jleechanorg/granola-exporter email-only pipeline — fetch Granola meeting notes via MCP, send ONE consolidated email per run to $USER@gmail.com ONLY when new notes exist, silent exit otherwise. No Google Docs, no Slack. Use when user says "granola…
原文の言語: 英語
Diagnose Hermes gateway health, Hermes monitor issues, and launchd service failures. Run when the user shares a monitor report, says "is everything ok", "check Hermes", "hermes is down", "slack-digest failed", "cronjob response failed", or when a launchd…
原文の言語: 英語
Audit which LLM models actually support the capabilities claimed in their config (code execution, structured JSON, tool calling) BEFORE promoting them to defaults, allowlists, or dropdowns. Covers Gemini code_execution, OpenAI reasoning_token footprint,…
原文の言語: 英語
Survey available LLM models across the project-app's configured providers (Gemini, Claude, OpenAI, Grok, Perplexity, Cerebras, etc.) and produce a cost-vs-quality upgrade matrix for a specific model upgrade request. Then dispatch an AO worker to land the…
原文の言語: 英語
Set up, verify, and maintain Claude Code + Codex routing through OpenRouter for a multi-machine pilot. Covers bash heredoc quoting for values with spaces/parens, cmux surface lifecycle for TUI picker overlay capture, alias install keeping…
原文の言語: 英語
Diagnose and safely reclaim disk space on macOS when a disk-usage alert fires (system or 3rd-party like `disk_usage_alert.sh`). Use when the user shares a disk-alert message, says "disk full", "low disk space", "running out of space", "free up disk", "reclaim…
原文の言語: 英語
Local HTTP proxy that lets Claude Code -p mode work with OpenRouter reasoning models (Kimi K3, DeepSeek R1, GLM 5.2, etc.). Strips thinking content blocks and buffers streaming responses so Claude Code's Node SDK doesn't drop stdout. ⚠ DEPRECATED 2026-07-21 —…
原文の言語: 英語
Route Claude Code and Codex through OpenRouter (Anthropic-Messages + OpenAI-Responses endpoints) directly without local proxies. Use when the user wants to test/use open models (GLM, Kimi, Qwen, etc.) or frontier models (Claude, GPT) under a single OpenRouter…
原文の言語: 英語
Wire Gmail + Slack notifications into a recurring launchd/cron job. Use when the user says 'send me a daily email', 'notify me on slack', 'send errors if something went wrong', 'email me the summary', 'ping slack on failure', 'daily recap', 'tag me when…
原文の言語: 英語
Diagnose "clean sweep" reports from scripted multi-agent review crons (Hermes `bug-hunt-daily.sh`, future siblings in the family — `wa-*-review-*`, `*-nightly-review-*`) when the report's PR/bug/agent-failure counts are all zero but the scan didn't actually…
原文の言語: 英語
Repair MCP Mail Slack bot scope and reinstall flow when post/mark behavior diverges after token or scope updates.
原文の言語: 英語
Diagnose why a Slack reply did not go to the right thread — failure modes 1-6 (5a-5h) plus direct-HTTP fallback. Includes wrong thread_ts from session header, deliver:local cron leak, cross-workspace bot-token block, home-workspace bot-not-in-channel, MCP…
原文の言語: 英語
Recurring cron-tick loop that watches a single Slack thread for a literal substring (for example WORKTREE APPROVED, MERGE APPROVED, DEPLOY APPROVED, or a custom approval token) and executes a pre-known callback (delete script, merge, deploy, etc.) on first…
原文の言語: 英語
Recover prior Slack MCP Mail bot setup from history, run browserclaw-guided OAuth/reinstall flow, and verify userscope/bot tokens.
原文の言語: 英語