Shape agent capabilities into your workflow.

Run Claude Code, Codex, and Aider sessions across many repositories and worktrees from a durable tmux-backed terminal UI. Fleet, split, babysit, and focus views keep waiting agents visible, while timelines, session search, and an optional private-network bridge extend supervision. It needs tmux and Git; native Windows support is not available.

tmuxlet runs coding CLIs inside tmux and gives scripts one blocking interface that returns text or JSON while preserving an interactive terminal. It normalizes Claude, Codex, Gemini, opencode, pi, Cursor, and Grok, uses an explicit completion signal, and reports completed, blocked, timeout, or exited runs with inspectable logs. A default 24-session cap and optional memory guard limit fan-out. Rust, tmux, and at least one target CLI are required.

py_ralph_frame provides a Python harness for spec-driven agent work: a plan defines the goal, specs define what to build, and each fresh-context iteration updates the implementation and project status. Pre-commit and pre-push gates can run linting, type checks, security scans, and tests for Claude, Codex, Gemini CLI, or Copilot; the harness intentionally does not create worktrees.

sub-agents-skills proposes a Markdown format for describing a sub-agent's role and skills so the same task-specific definition can be reused with Codex, Cursor, Gemini CLI, Claude Code, and related tools. It is useful for keeping repeated agent roles in readable files. Actual loading and execution still depend on how each target assistant supports Markdown or skill files.

This skill collection gives Claude Code package-oriented guidance for more than 250 R packages across data wrangling, visualization, machine learning, statistics, bioinformatics, spatial work, web applications, and package development. After installation, users can request R analysis or code in natural language; R 4.0 or newer is required.

This skill gives AI agents a structured framework for local SEO, Google Business Profile work, map-pack rankings, reviews, and visibility across ChatGPT, Gemini, Grok, and Google AI surfaces. It distinguishes map visibility through SoLV from AI mentions through SAIV; with the Local Falcon MCP it can analyze live data and scans, while without MCP it provides guidance and best practices.

Pear is a hosted MCP integration for Claude Code and OpenClaw that connects to iCloud Calendar, Reminders, Contacts, and Mail. It exposes scheduling, contact lookup, reminder management, and mail operations through natural-language requests, with support for macOS, Linux, and Windows. Setup requires a Pear account, an Apple app-specific password, and an API key; the project says connected data is proxied in real time rather than stored on Pear’s servers.

Scholar Sidekick MCP Server resolves DOIs, PubMed IDs, ISBNs, arXiv IDs, ISSNs, ADS bibcodes, and WHO IRIS URLs into formatted citations or bibliography files. It supports batch input, more than 10,000 CSL styles, nine export formats, retraction checks, and open-access lookup. The public API can be used anonymously at a rate-limited tier, while clients may instead connect to the hosted Streamable HTTP MCP endpoint.

Ask a product-management question and the skill selects five relevant experts from an index of 302 Lenny’s Podcast guests. It reads their full interview transcripts in parallel, presents each perspective, and synthesizes shared themes, disagreements, and actionable takeaways with citations and timestamps. Local transcript files, Python 3.9 or newer, and Claude Code are required.

Atomic Emails guides a SaaS campaign through a structured five-step method built around one email per idea. It gathers the audience, campaign goal, success steps, obstacles, positioning, and existing resources, then turns that material into a storyboard and finished copy. The workflow fits onboarding, trial expiry, customer success, lead nurture, and re-engagement campaigns, and can start from an existing brief or storyboard.

Agent Tutor turns Claude Code into a tutor that requires learners to explain, apply, and transfer a concept before marking it understood. Zero-hint quizzes expose weak recall, while FSRS schedules review separately for each concept. Progress, misconceptions, and session history persist in local Markdown files, and lessons can be grounded in PDFs, documents, web pages, or source code.

Six-Layer Product Research guides Claude Code through a structured problem-definition process before solution work begins. It examines customer language, ecosystem perspectives, behavioral frictions, failed status-quo attempts, new technology enablers, and the changes a team can realistically influence. Separate findings are then synthesized into a multi-perspective definition that helps product teams test assumptions and understand why earlier approaches failed.

The skill searches 20,972 Arabic-English records from the Quran, Sahih al-Bukhari, and Sahih Muslim through Supabase and pgvector. It supports exact lookup, topical search, cross-referencing, and checking whether a report appears in the two Sahih collections, while presenting returned text with citations instead of reconstructing it from memory. It does not issue fatwas and requires a Supabase deployment plus embedding setup.

Search popular Xiaohongshu notes by keyword for topic ideas, then publish an image-and-text post with a title, body, images, tags, and visibility settings. Posts can be scheduled from one hour to fourteen days ahead, and QR-code login saves the session for reuse or switching accounts. Python 3.8 or later and the xiaohongshu-mcp binary are required.

Designed for teams packaging ERP, CRM, HIS, and similar workflows, the tool probes each step for data sources, decision logic, and failure handling before generating the skill. External interfaces are wrapped in Python scripts with mock support, followed by evaluation and iteration. It also supports editing existing skills, changing deployment parameters, and importing reusable domain knowledge.

Darce is an account-based terminal coding agent that can read, write, and edit files, search a codebase, run shell commands, and fetch web content after installation and login. It supports in-session model switching or automatic routing, plus context compaction, resumable sessions, and cost tracking. The free plan limits monthly requests and exposes only a subset of read-oriented tools.

This static Agent Skill helps an assistant advise you on recording a voice performance for conversion into instruments such as piano, guitar, violin, drums, or saxophone. It covers microphone and environment choices, instrument-specific technique, prompt writing, file formats, and diagnosis of muddy, robotic, off-rhythm, or clipped output. The skill supplies guidance only; it contains no executable conversion code, telemetry, or backend calls.

pmstack lets product managers build an evaluation lifecycle inside Claude without starting from an ML engineering dashboard. Its commands turn observed failures into task suites, graders, pass bars, and runnable YAML, then score a real target, review failed transcripts, and watch for regressions after model changes. PRDs, customer-signal synthesis, briefs, and launch checks are included, but the differentiator is making AI evals directly usable by PMs.