claude-skills

Use when delegating coding, review, analysis, testing, or repo-maintenance work to OpenAI Codex CLI from Claude Code or another agent, especially when the user asks to run codex, get a Codex review, use codex exec, save Codex output, stream JSON events, or run autonomous codebase work in a local repository.

Use when the task is to FIND ideas, solutions, root causes, workarounds, prior art, or benchmarks ONLINE — researching a problem/idea/solution on the web — NOT a simple single-doc API lookup. Searches across MANY languages in parallel: spawns one research subagent per language (English, Chinese Simplified, Chinese Traditional, French, Russian, Hebrew, Spanish), each searching native technical terms and native forums, then synthesizes. Triggers: "research this", "search online for", "find a solution/idea/approach", "how do people solve", "is there prior art", "find workarounds", "deep multilingual web research".

Use when the user says "bg-it" or "spawn this in background" or "run this as bg agent" — spawns a background claude session for ANY prompt type or archetype, including interactive ones. Claude Code only.

Micromanager (`/mm`) - Living PM skill that maintains continuous HANDOFF.md task state, repo-level PM dashboards, scheduling fields for target dates and risk, implementation prompts, file inbox updates from agents and CEO recommendations, durable insights, and task close-out. Auto-detects fresh vs continuing vs update vs dashboard mode; never writes production code itself; preserves the Section 0 continuation contract.

Send text messages and files to a configured Telegram chat (group, channel, or DM with bot). Supports forum/topic groups. Reads bot token + chat ID from ~/.claude/.env so the skill itself contains no secrets.

Hybrid LLM-engineering research — synthesizes a podcast-grounded answer from the Langtalks corpus with current web search results. Triggers on /langtalk, "langtalks", "what does langtalks say about", and LLM-system architecture/trade-off questions where named-practitioner provenance matters.

Use when a frontend development phase is complete and needs visual verification. Triggers on any browser-based project (React, Vue, Next.js, Angular, plain HTML) after building screens, completing a spec section, or when visual regressions need catching. Runs fully autonomous by default.

Apply RTL Hebrew rules when generating ANY document containing Hebrew text. Use alongside pptx-generator, minimax-docx, minimax-xlsx, minimax-pdf, or any HTML/web output. Fixes BiDi confusion, wrong punctuation, reversed parentheses, layout mirroring, and comma placement errors. Triggers: Hebrew, עברית, RTL, right-to-left, rtl document, Hebrew presentation, Hebrew Word doc, Hebrew Excel, Hebrew PDF, Hebrew slides, ישראל, עברי.

Convert an approved PRD into a phase-gated implementation SPEC. User-facing entry point that dispatches to the prd-to-spec babysitter process via /babysitter:call (interactive) or /babysitter:yolo (auto). Stack-agnostic.

Convert a raw task (tracker ticket, email, or text) into a fully characterized PRD. User-facing entry point that dispatches to the task-to-prd babysitter process via /babysitter:call (interactive) or /babysitter:yolo (auto). Stack-agnostic.

Deploy a self-contained HTML file to Firebase Hosting with HMAC token-gated access. Pages are stored in Firebase Storage (GCS) and served by a Cloud Function. Access URLs include a derived token so pages aren't publicly indexable. Use this skill when asked to: deploy HTML to Firebase, share a page as a live link, host an interactive dashboard/page, or set up Firebase HTML hosting. Two phases: SETUP (one-time per project) and DEPLOY (per page).

Use when asked to find gaps between a plan file and the actual codebase. Spawns parallel Claude subagents (one per codebase area: backend, frontend, tests, infrastructure, scripts) each cross-referencing the plan against real code. Aggregates into a structured gap report: planned-but-missing, implemented-but-not-planned, and partial implementations. Works in any git repo.

Autonomous GitHub issue fixer. For each open issue: Session 1 writes a spec + runs /deep-verify-plan (≥95/100), Session 2 uses /writing-plans to produce a TDD task list, Session 3 implements with TDD + /verification-before-completion then commits and pushes. Quality gates validate each artifact. Loops until no open issues remain. All orchestration events are visible inline in the CC terminal.

Universal coding standards for any language or stack. Run BEFORE writing code (to set intent) and AFTER writing code (to verify quality). Covers naming, structure, security, tests, and commits.

Coding standards, patterns, and conventions for TypeScript/NestJS/React/MUI/Mongoose projects. Apply when writing or reviewing code in this stack.

Use when asked to verify, improve, or sharpen a plan file using deep plan verification. Runs the deep-plan-verification babysitter process (yolo mode) on a plan file — no coding, no implementation, only iterative plan QA. Shows before/after diff and final quality score at the end.

Use when asked to audit code for bugs without fixing them. Scans the repo with 3 independent expert judges (weighted scoring + expert veto), proves root causes, and generates a report. NEVER modifies code — read-only audit.

Use when asked to hunt bugs, find and fix bugs, audit code quality, or run /bh. Launches the generic/bug-hunter babysitter process on the current repo. Supports yolo (auto-fix) and interactive (breakpoints) modes.

Use when asked to create a multi-session babysitter script, run multiple babysitter sessions sequentially, chain babysitter runs, or generate a run-sessions.sh file

Use when the user wants to run continuous autonomous bug hunting with TDD and convergence scoring until the codebase reaches production quality. Triggers on "bh forever", "hunt bugs forever", "fix everything until done", "converge to quality", or any request for a never-ending self-improving bug fix loop.