Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
leo-pe2/google-health-hermes-agent-mcp
Skill Claude CodeCodex
Use when answering questions about the user's health, sleep, recovery, activity, workouts, or physiological responses, especially when combining Google Health data with calendar events, tasks, notes, weather, or durable user preferences. Query the local Google Health MCP precisely, check freshness, compare against…
Skill Claude CodeCodex
Linux/X11 desktop control — inspect, click, type into visible windows via AT-SPI + xdotool. Returns @eN refs from the focused window's accessibility tree.
Skill Claude CodeCodex
Manages AIREADME.md files so AI agents remember your project conventions across every session.
Skill Claude CodeCodex
OpsContext for AI Agents — the ops + compliance layer Claude Code can't grow natively. Read-only fleet visibility (PM2/nginx/Docker/git/cron) + tamper-evident hash-chained audit log producing evidence aligned with SOC 2 CC7.2 and ISO 27001 A.12.4.1 (evidence artifacts only — OpsContext itself is not certified; see…
Skill Claude CodeCodex
give your ai agent a phone number. provision lines, make calls, check compliance, all from claude code. saperly is the phone carrier for ai agents.
Skill Codex
Operate hosts and Agent OS services through the governed systems-manager MCP provider and GraphOS delegation. Use for OS and package lifecycle, services, processes, network and disk inspection, storage or BMC health, firewall and managed-file work, agent identity, specialist registry, scheduler, watchdog, host…
Skill Claude CodeCodex
Analyze a chart — set up symbol/timeframe, add indicators, scroll to key dates, annotate, and screenshot. Use when the user wants technical analysis or chart review.
Yongthyuan/agent-context-bridge
Skill Claude CodeCodex
Continue CC Switch-managed Claude Code, Codex, Gemini CLI, opencode, Cursor, and Aider sessions through Agent Context Bridge handoff packets.
Skill Codex
Use when importing resume JSON through MCP fails due to escaping/encoding corruption (for example double-escaped quotes and backslashes). Runs a local safe-import command that normalizes JSON first, then writes section-by-section to the remote resume MCP over SSE.
Skill Claude Code
Local, API-key-free persistent memory over MCP (SQLite + FTS5, no LLM/embeddings). Use when the user asks to remember, recall, or search something persistent across sessions ("hafızaya al", "hatırla", "bunu unutma", "remember this", "what did I say about..."), or wants to set up/troubleshoot the memorry MCP server…
Skill Claude CodeCodex
Uses TracePulse MCP server for backend runtime error monitoring. Use after ANY backend code change to check for errors, verify fixes, and triage crashes. TracePulse watches dev server logs (stdout/stderr or log files) and exposes parsed, scored errors as MCP tools. Works with any language - Node.js, Python, Go, Java…
Skill Claude CodeCodex
A set of tools for creating, inspecting, changing, and simulating MATLAB Simulink models. Simulink is a visual environment for building and testing systems from connected blocks.
Skill Claude CodeCodex
Use whenever you need to check on, read from, write to, or monitor a WAGO PLC (PFC200, PFC300, CC100, CC100-IEC62443, Edge Controller, WP400, TP600) via the wago-plc MCP server - plain-English chat requests ("what's the firmware on 192.168.1.10?", "turn on NTP sync", "watch the LED every 30 seconds", "list my PLCs")…
Skill Claude CodeCodex
Use whenever a .pdf file is involved — as input, output, or both: reading or extracting text and tables from a PDF; merging, splitting, or reordering pages; rotating, cropping, or deleting pages; adding watermarks or stamps; extracting images; creating a new PDF from scratch; or OCR'ing a scanned PDF to make it…
Skill Claude CodeCodex
Use when an agent needs to automate live Microsoft Excel workbooks through Open Workbook MCP with the public excel.agent.run workflow: inspect workbooks, read/write ranges, update tables, preserve templates, repair formulas/styles, create pivots/charts, validate reports, save/export files, coordinate multiple agents…
humanjesse/openclaw-thunderbird
Skill Claude CodeCodex
Search, read, compose, reply, and forward emails via Thunderbird.
Skill Claude Code needs its repo
AI agent guardrails — defends prompts against injection attacks, jailbreaks, and evasion; scans LLM responses for leaked secrets and harmful content. Up to 36 inline defenses (24 default), 3 output scanners, and adaptive ranking. Keywords: shield, guardrails, prompt injection, defense, security, scan.
evergreen-ci/evergreen-mcp-server
Skill Claude CodeCodex
Complete reference for every Evergreen MCP tool — what it does, when to use it, parameters, return values, and how results chain into other tools.
Skill Claude CodeCodex
A code-review workflow used with a local code-review server. It examines files, uncommitted changes, or an entire project for common quality and security problems.
Skill Claude CodeCodex
Skills are higher-order, multi-step workflows that compose atomic graph tools into reliable, end-to-end Blueprint creation sequences.
Skill Claude Code
Creates and maintains Figma Code Connect template files that map Figma components to code snippets. Use when the user mentions Code Connect, Figma component mapping, design-to-code translation, or asks to create/update .figma.ts or .figma.js files.
Skill Claude CodeCodex
Plan MMS FGM/MEC/EDP/SCM/FPI/HPCA route-scout and overview workflows as Agent Kit skill/resource recipes without expanding the default MCP tool surface.
Skill Codex
A set of instructions for using MiMo Bridge MCP, a local connection that lets a coding agent delegate tasks to MiMo Code or Reasonix. It covers starting, waiting for, retrying, reviewing, and safely managing those tasks.
Skill Claude CodeCodex
Build, edit, connect, and deploy premium websites with the Sophia Stack — an AI-native full-stack framework. Use when creating or editing a site as a compact Site Model (blocks + style preset + effects + backend connections) instead of hand-writing JSX/CSS. Optimized for minimal token cost, real-time edits, and safe…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: