Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

awesome-hermes-agent

A curated list of skills, plugins, tools, integrations, and resources for Hermes Agent by Nous Research
https://github.com/Anil-matcha/awesome-hermes-agent

Last synced: 15 days ago
JSON representation

  • Detection & Media Forensics

    • Deployment

      • beta - ai](https://github.com/resemble-ai) — Deepfake detection and media safety for agents. Detects AI-generated audio, images, video, and text; traces audio source (ElevenLabs, Resemble, etc.); applies invisible watermarks for provenance tracking; verifies speaker identity (Beta). Core principle: never declare media real or fake without a completed detection result.

  • Domain Applications

    • Deployment

      • experimental - improving robotics via VLA model fine-tuning. Applies the Hermes learning loop to physical robot control.
      • beta
      • experimental
      • beta
      • beta
      • beta
      • experimental
      • experimental
      • beta - ready kits from startup ideas — market analysis, pitch deck, and financial projections.
      • beta - time US$10 builds and runs the chosen idea. Open source (MIT), no subscription, no signup.
      • beta - chain blockchain cash flow analyzer with a WebGL dashboard. On-chain forensics and flow visualization.
      • beta - native open-source ERP and accounting you self-host and run in plain English. Invoicing, inventory, general ledger, payroll, and multi-company books in one shared database.
      • production - Automation](https://github.com/Strategic-Automation) — Supervised, agentic Hermes pentest profile: 31 methodology playbooks (OWASP Top 10, API Top 10, LLM Top 10), interactive scoping + approval gates, optional `violin-guard` plugin. Installs with `hermes profile install https://github.com/Strategic-Automation/violin`. Hermes-native, no extra API keys.
      • beta - hosted investment decision runtime. Four independent LLM roles (Macro Strategist, Quant Analyst, Risk Officer, CIO) debate any yfinance symbol (US / HK / A-share / ETF / crypto / commodities) and return a calibrated BUY/HOLD/SELL verdict — never places real orders, local SQLite ledger only. Install: `hermes plugins install longsizhuo/openInvest --enable`; MCP backend via `uvx openinvest mcp`.

  • Forks & Derivatives

    • Deployment

      • beta - critical deployments requiring formal trust verification.
      • experimental - ramp for Oracle environments.
      • beta - deployed Hermes with pre-configured infrastructure templates. Skips local setup entirely.
      • experimental - world tasks for producing fine-tuning data at scale.

  • Guides & Documentation

    • Deployment

      • beta
      • production - by-step WSL2 Ubuntu setup guide for running Hermes on Windows.
      • beta - maintained wiki with practical patterns and deployment advice for building autonomous agents with Hermes.

  • Integrations & Bridges

    • Deployment

      • beta
      • beta
      • beta - founder.
      • production - io) — Long-term memory layer for agents with retain/recall/reflect workflows. Integrates with Hermes via plugin or MCP and supports semantic, graph, and temporal retrieval.
      • beta - hosted Honcho memory backend for Hermes. Useful when you need stronger cross-session memory behavior with full local control.
      • beta - native memory provider for YantrikDB. Self-maintaining memory: `think()` canonicalizes duplicates, `conflicts()` surfaces contradictions instead of silently overwriting, and every `recall()` result carries a `why_retrieved` reason list for explainable ranking. Drop-in install, talks HTTP to a Rust backend.
      • beta - Persistent memory for AI agents over MCP. Runs as a remote HTTP server with OAuth at https://mcp.mnemoverse.com/mcp, or as a local stdio server via `npx -y @mnemoverse/mcp-memory-server` with an API key: the two kinds of MCP server Hermes `mcp_servers` takes. Recall is re-ranked by feedback on whether a recalled memory helped or misled. MIT server, hosted engine, free tier.
      • experimental
      • beta
      • beta - agent delegation skill bridging Hermes, Codex, and Claude Code. Routes subtasks to the best-suited agent automatically.
      • experimental - chain analytics and wallet data to Hermes via MCP.
      • experimental - perspective council MCP server. Multiple AI viewpoints debate before the agent commits to a decision.
      • production - friendly sites with agentic scoring, category filters, and live JSON-RPC verification. Wire it into Hermes to let the agent find and evaluate new tools on its own.
      • experimental - aware routing layer. Routes compute-heavy tasks to available GPU infrastructure.
      • beta - Girgis](https://github.com/Andrew-Girgis) — Full Outlook/Hotmail/Microsoft 365 integration via Microsoft Graph API. Email, calendar, contacts, user profile, and free/busy scheduling. OAuth2 with auto-refresh. Includes a preview-before-send pattern that prevents shell `$` variable mangling when agents compose emails containing dollar amounts.
      • beta - first Android control for Hermes over WiFi — no USB, ADB, or root required once the AIVane service is running. Supports health checks, app listing, UI tree inspection, taps, text input, swipes, navigation, screenshots, and inspect→act→smoke flows. Only connects to user-provided device URLs.
      • beta - time WebSocket messaging, shareable profile cards, and local/web inbox. All actions require explicit user request. Bilingual (English + Chinese).
      • beta
      • production - bound, self-custodial x402 payment wallet via `npx -y @piprail/mcp`. Lets Hermes autonomously pay 402 "Payment Required" URLs across EVM, Solana, TON, Tron, NEAR, Sui, Aptos, Algorand, Stellar, and the XRP Ledger — funds settle straight to the merchant, no facilitator, no fee, capped by a spend policy the model cannot exceed. MIT.
      • production - pay any number of recipients in a single on-chain transaction across 13+ chains (Base, Ethereum, Solana, Polygon, Arbitrum, and more). 151 paid primitives covering escrow, token bridges, gas estimation, oracle data, and AI inference — all non-custodial, pay-per-call via the x402 micropayment protocol. Available as a [Hermes optional skill](https://github.com/NousResearch/hermes-agent/pull/49716) (`hermes skills install official/blockchain/spraay-x402`) or via [MCP server](https://smithery.ai/servers/Plagtech/Spraay-x402-mcp) with 161 tools. Batch contract verified on Base: `0x1646452F98E36A3c9Cfc3eDD8868221E207B5eEC`.
      • production - compatible API (BYOK), or fully offline ComfyUI. README ships a tested Hermes `mcp_servers` YAML with the timeout overrides needed for video generation workflows. 1k+ stars.
      • production - per-call data capabilities via x402 MCP: stocks, crypto, DeFi, on-chain analytics, geopolitical signals, and Polymarket intelligence. USDC micropayments on Base — no API keys or accounts required.
      • production - text and audio history for Hermes recall, meeting context, and work summaries through [MCP](https://github.com/screenpipe/screenpipe/tree/main/packages/screenpipe-mcp#installation). Capture is local by default; configured cloud features, remote clients, and model providers can process context off-device. Source-available under the Screenpipe Commercial License.
      • beta - Tag-style context-selection layer for Feishu/Lark and Slack group chats (DingTalk supported, mention-only). Answers only when @-mentioned, selecting bounded evidence — the original images, your notes, and the relevant replies — instead of full-history RAG. Per-chat memory isolation with redacted admin audit and lifecycle controls.
      • beta - to-end encrypted inbox via `--deliver arelay`. Reports and media arrive in a private web inbox you unlock with a passkey, instead of scrolling gateway logs. MIT licensed.
      • beta - platform) — Single Responses-style HTTP/SSE API that routes chat turns to a local Hermes worker or an OpenClaw backend, with reconnectable stream replay
      • beta - duplex Discord voice bridge to the Gemini Multimodal Live API, with 40+ function tools, vision, per-user memory, and mid-call delegation to Codex/OpenCode. Standalone plugin, MIT.

  • Multi-Agent & Swarms

    • Deployment

      • experimental - agent swarm framework. Coordinates multiple agents toward shared goals with distributed task execution.
      • experimental
      • beta - native SDLC with Hermes and Supermodel code graph. Full software development lifecycle run by coordinated agents.
      • beta

  • Official Resources

    • autonovel - form writing pipeline built on Hermes. Generates full manuscripts (100k+ words) end-to-end using the agent loop.
    • hermes-paperclip-adapter
    • hermes-agent-self-evolution - improvement using DSPy and GEPA (Genetic Evolution of Prompt Architectures). The research pipeline for optimizing Hermes's own prompts and behaviors.
    • tinker-atropos - tuning tool-calling models on real agent trajectories.

  • Operational Playbooks

  • Skills & Plugins

    • agentskills.io Ecosystem

      • production - improving QA agent with SKILL.md modules for authoring natural-language web/mobile tests, debugging failed runs, and triaging results. Works with Claude Code and Codex-style skill loaders.
      • production - in `youtube-content`. Routes through TranscriptAPI's backend (15M+ transcripts/month) to stay unblocked on cloud IP ranges. 12 sub-skills covering the full search → fetch → bulk extraction flow.
      • production
      • production
      • production - forest-labs) — Official FLUX model skills for image generation. First-party skills from the team behind FLUX.
      • production - safe schema validation to agent skill inputs and outputs.
      • beta - running task execution with proper lifecycle management — progress tracking, checkpoints, and failure recovery.
      • beta - step skill execution into observable pipelines.
      • beta - compliant skills. Input a codebase, get installable skills.
      • beta
      • experimental - rip](https://github.com/KYC-rip) — Monero (XMR) blockchain gateway for agents. Enables private cryptocurrency transactions from agent workflows.
      • beta
      • beta
      • beta - Systems](https://github.com/Merit-Systems) — Gives agents access to 300+ premium APIs through a single skill backed by a wallet balance for x402 or MPP payments. Covers web scraping, image generation, email delivery, and more.
      • beta - dev](https://github.com/Xquik-dev) — Typed X (Twitter) access across 43 SKILL.md folders covering reads (search, timelines, mentions, trends, articles, bookmarks, for-you), writes (post, DM, follow, profile updates), bulk extraction (followers, communities, lists, spaces), and AI composition (write-tweets, threads, optimize, going-viral). Typed JSON in/out through the Xquik API — no browser automation.
      • production - ai](https://github.com/Agents365-ai) — Generates draw.io diagrams from natural language and exports to PNG/SVG/PDF. SKILL.md format, works across Claude Code, OpenClaw, Hermes, and Codex. 1.1k+ stars.
      • production - io](https://github.com/nexu-io) — Local-first, open-source design alternative with 31 composable skills over 129 design systems, plus image, video, and audio generation. Integrates with Hermes via ACP/JSON-RPC. 28k+ stars.
      • beta - ai) — Cross-agent skill for running AI image, video, music/audio, and model API jobs from CLI-capable agents.
      • beta - Agent skill for operating Sequenzy lifecycle email marketing and transactional email workflows: subscribers, segments, campaigns, sequences, templates, and stats.
      • production - time quotes, K-line charts, fundamentals, portfolio analysis, technical analysis, options/warrants, watchlist, and earnings reports for HK/US/A-share/SG markets. Trilingual (Simplified Chinese / Traditional Chinese / English). Install: `npx skills add longbridge/skills -g`.
      • production - labs) — Job-search and career-discovery skill. Searches tech companies by their actual tech stack, stars the ones that match the user's stack, pulls jobs from that curated shortlist, and reads/writes the user's developer profile through conversation. Backed by a remote MCP server (`https://api.remoet.dev/mcp`, Bearer API key or OAuth). One SKILL.md that runs unchanged on Hermes, OpenClaw, and any MCP-capable harness. Free tier.
      • beta - to-agent social network as an agentskills.io skill. The `tinyplace` CLI lets an agent claim an @handle identity, get discovered in an open directory, message peers over Signal end-to-end encryption, and transact in USDC and SOL on Solana via x402. One SKILL.md that runs on Hermes, OpenClaw, and any agentskills.io-compatible harness. [ClawHub](https://clawhub.ai/tinyhumansai/tinyplace)
      • beta - to-end encrypted inbox, and submits outbound email drafts for human approval before send. Works on any agentskills.io-compatible host. MIT licensed.
      • production - focused agent skills on the agentskills.io standard: draft posts and comments, extract hook formulas, audit drafts, reduce AI tells, optimize a profile, and plan a weekly cadence. Each is a `skills/*/SKILL.md` folder that Hermes loads natively; `hermes claw migrate` imports them from an OpenClaw setup. MIT licensed.
      • beta

    • Community Skills

      • beta - evey](https://github.com/42-evey) — Plugin suite (195+ ⭐): Discord voice bridge with the Gemini Live API, a WhatsApp (Baileys) bridge, plus goal management, inter-agent bridging (Claude Code ↔ Hermes), model selection, and cost control.
      • beta - skill that auto-generates reusable skills from your workflows. Aim it at a repetitive task and it creates a skill for it.
      • beta
      • experimental - themed skill pack. Wraps real development operations (refactoring, linting, testing) in a tabletop RPG interface.
      • experimental - reasoning layer where the agent generates stronger prompts before acting.
      • experimental
      • beta - healing. Monitors services, diagnoses failures, and applies fixes. Works naturally with Hermes's cron scheduling.
      • beta - improvement system that tracks agent performance, identifies weak skills, and iterates on them automatically.
      • beta - as-speaker. The only Linux-native Spotify skill in the ecosystem.
      • experimental
      • experimental
      • production - platform coding-agent skillpack for smaller diffs, evidence-led debugging, token control, browser verification, and reviewable `SKILL.md` workflows with validation and static security audit.
      • beta - vinc](https://github.com/adnw-vinc) — Self-hosted Nextcloud bridge covering files (WebDAV), notes (Nextcloud Notes API), calendar and tasks (CalDAV), and contacts (CardDAV). App Password auth, configurable timezone, guided setup. Fills the self-hosted-cloud gap for Hermes users running their own infrastructure.
      • beta - agent orchestration skills for Hermes, inspired by `oh-my-claudecode` and rebuilt on Hermes primitives. Covers deep-research, deep-interview, `ralplan` (Planner → Architect → Critic consensus), `ralph` (execute → verify → iterate), `triage`, and `autopilot`. Composes end-to-end: research → interview → plan → verified execution.
      • production - Labs](https://github.com/Infrasity-Labs) - Open-source, cross-platform agent skills for Claude Code and agentskills.io-compatible platforms. These skills are for SEO, GEO (Generative Engine Optimization), AI discoverability, and developer marketing.
      • production - source analytics skills. Turnkey operator agent for running a business from Hermes.
      • beta - ai](https://github.com/loremaster-ai) — AI scrum-master / PM skill pack extracted from a production multi-project deployment: sprint ceremonies with planning poker, backlog proposals, plan-conflict detection, and a per-project wiki vault with a knowledge graph. Every external write goes through per-item human approval buttons.
      • beta - platform AgentSkills-compatible skills for the Currents News API, with installable packages for Hermes and other agent runtimes.
      • beta - studio) — Host-agnostic `SKILL.md` workflows for SEO, GEO, Google Ads, and Meta Ads, with MCP connectors for live account analysis and approval-gated changes.
      • beta - court) — Agent-to-agent commerce skill: natural-language mandates, ERC-7710 delegated permissions, x402 payments, escrow, and dispute resolution. Install: `hermes skills tap add internet-court/internet-court-skill`.
      • experimental

    • Plugins

      • beta - hosted service](https://github.com/eloktev/artifact-relay), with separate publishing and viewing credentials and sandboxed HTML rendering.
      • production - agent team plugin. 8 role templates, event-driven discussions with voting, heartbeat-driven project management, kanban task dispatch, AGENTS.md shared context, dashboard UI. 17 tools, 3 hooks.
      • beta - ai](https://github.com/plur-ai) — Shared memory layer for AI agents using an open engram format (YAML). Useful for persistent learning patterns across Hermes sessions.
      • experimental
      • beta - cla](https://github.com/robbyczgw-cla) — Multi-provider web search that intelligently routes across Serper, Tavily, Exa, and more. Better result quality and source diversity than the built-in search.
      • beta - grade weather plugin with NWS model imagery, NEXRAD radar, and meteorological calculations.
      • beta - hosted realtime voice gateway and Dashboard plugin for Hermes, with background run progress and reconnect-safe completion notices. Supports local speech-to-speech on Apple Silicon, Gemini Live, and OpenAI Realtime.
      • experimental
      • production - selected tabs through an authenticated loopback broker.
      • experimental - account testing or browsing with different saved sessions.
      • experimental
      • beta - evey](https://github.com/42-evey) — Claude Code plugin for bridging with Hermes. Lets Claude Code and Hermes share context and hand off tasks to each other.
      • beta - Analytics](https://github.com/Agent-Analytics) — Native Signals dashboard tab for Hermes with read-only multi-project analytics, explicit timeframe controls, and theme-aware UI.
      • beta - dev](https://github.com/Xquik-dev) — Native Hermes Agent plugin for X/Twitter search, read, and guarded action workflows through Xquik. Ships plugin manifests, SKILL.md bundles, installer prompts for API keys, and compatibility tests against current Hermes plugin behavior.
      • beta - driven companion to v0.12's built-in Curator. Observes tool and skill events into a local SQLite store, backfills existing session history, generates reports and dry-run proposals, and runs a guarded daily loop that only appends low-risk evidence-backed notes with backups and rollback manifests.
      • beta - style Dynamic Workflows to Hermes: model-written sandboxed scripts coordinate subagents with `agent`, `parallel`, and `pipeline`, supporting up to 1000 subagents, persisted/resumable runs, a live `hermes-workflows` dashboard, and `/workflows` history.
      • beta - ai/rtk) before it reaches the LLM context. 60–90% token reduction on shell commands. Zero config — auto-loads on gateway boot.
      • beta - backed secret and key lifecycle, preflight plugin scanning and enforcement, MCP tool auditing, cost governance, diagnostics, image routing, and remote assistant delegation. 7 tools + 2 hooks via native Hermes plugin manifest. 201 tests, MIT.
      • experimental - 688](https://github.com/jim-688) — Architecture governance framework for Hermes: layered memory (Hot/Cold/Config/Runtime), evidence-driven iteration (Hypothesis/Proposal/Observation), and capability-driven design. Documentation repo with runnable demo.
      • beta - in dangerous-command guard with user-defined patterns, works with the standard approve-once/session/always/deny flow, and ships `hermes custom-dangerous-patterns` CLI commands to manage them.
      • beta - receipt status. Adds /cron-status.
      • experimental - spirit-realm](https://github.com/the-spirit-realm) — Hermes desktop plugin: sidebar fleet page for local bots (profiles) plus per-bot Home dashboards. Install: `hermes plugins install the-spirit-realm/hermes-bot-hq`.

    • Skill Registries & Discovery

      • beta
      • production - platform skills marketplace.
      • production - st](https://github.com/supernovae-st) — Deterministic workflow runner Hermes can delegate to: repeatable AI jobs as reviewable `.nika.yaml` files, checked before a token is spent and tamper-evident after. Ships a skill pack plus a read-only MCP oracle; the offline mock path needs zero keys.
      • production - protocol agent discovery across MCP, A2A, and agents.txt. Searchable directory of 18K+ MCP servers and agents.
      • production

  • Tools & Utilities

    • Deployment

      • beta
      • beta - ready Docker image for cloud Hermes deployments. Infrastructure wiring pre-configured.
      • experimental
      • experimental
      • beta - hosted VPS hosting options across OpenClaw and Hermes Agent.
      • production - Ray-Martin) — Managed Hermes hosting on Google Cloud (seaotter.dev, not seaotter.ai): isolated agents, MCP from Cursor/Claude/Codex, 14-day Hobby trial then $69/mo, no VPS or SSH.

    • Skill Registries & Discovery

      • beta - reach](https://github.com/dodo-reach) — Native macOS workspace for Hermes built around a direct, host-first SSH connection model. Includes a real embedded terminal, session browsing, file editing, skills browsing, and usage metrics — no extra gateway or daemon.
      • production - labs](https://github.com/builderz-labs) — Open-source dashboard for AI agent orchestration. Manage fleets, dispatch tasks, track costs, and coordinate multi-agent workflows. Self-hosted, SQLite-powered. 3.7k+ stars.
      • experimental - Flow](https://github.com/Tranquil-Flow) — Terminal neurovisualizer with 42 animated themes. Decorative overlays for agent activity.
      • beta - labs-ai) — Static analysis for AI agent configs, tool descriptions, and system prompts, returning a PASS / REVIEW / FAIL verdict. Zero-LLM, deterministic checks, built for CI.
      • beta - AI-Corp) — Records agent runs for byte-for-byte offline replay and lets you fork a run from a checkpoint onto another model. Supports Hermes through its process and socket boundary adapter. Apache-2.0.
      • beta - source CLI and local MCP bridge with a Hermes adapter for routing multimodal generation across 2,000+ models and APIs. Installs a Hermes Skill and MCP entry, with backup and rollback safeguards. Apache-2.0.
      • beta
      • beta - hosted control plane for deploying and operating Hermes and OpenClaw fleets on Docker and Kubernetes, with lifecycle management, monitoring, terminal and log access, integrations, and CLI/MCP interfaces.
      • beta
      • experimental - tyler](https://github.com/unmodeled-tyler) — AI-native Linux browser with MCP control and autonomous browsing. A full browser built for agent use, not just a headless wrapper.
      • production - inc](https://github.com/jo-inc) — Stealth headless browser server with a REST API that bypasses Cloudflare, bot detection, and anti-scraping. Drop-in Puppeteer/Playwright replacement used in Hermes's own browser automation. 4k+ stars, MIT. The right backend when your VPS-hosted Hermes keeps getting blocked.
      • beta
      • beta - llama](https://github.com/pyrate-llama) — Single-file glassmorphic web UI with SSE streaming, tool call visualization, activity panel, image analysis (Gemini Vision), session management, PDF export, skill browser, memory viewer, and mobile support. Python proxy on port 3333, no build step required.
      • beta - workspace, focused on ops.
      • production - time streaming chat across multiple backends, unified config for 8 messaging channels, token/cost analytics with 30-day trends, cron scheduling, multi-profile gateway management, multi-agent group chat with @mention routing, and a WebSocket terminal. 3.6k+ stars.
      • beta - evey](https://github.com/42-evey) — One-command setup for the full Hermes stack with free models and 29 pre-configured plugins.
      • beta - loads relevant context before queries reach the agent.
      • beta - hosted decision memory that stores decisions **and their rejected alternatives** as a structured field, then surfaces the veto with its recorded reason when a task revisits a ruled-out approach. Passive capture through a Perception API on your own Postgres; always-on decision summary, per-turn retrieval, and a `robrain_search` tool, all fail-open. Optional fully-local LLM via Ollama/LM Studio, Apache-2.0. Installs with `npx robrain install --hermes`. Ships VetoBench, a reproducible benchmark of rejected-approach recall with committed run receipts.
      • production - ML](https://github.com/AMAP-ML) — Open-source companion that auto-evolves, deduplicates, and improves your skill library from real session data. Adds a post-task evolution loop on top of Hermes's built-in skill creation. Native integration via `~/.hermes/skills`. 705 stars, MIT licensed.
      • beta - first TUI/CLI for post-run session audits covering cost and token spikes, tool failures, retry loops, latency gaps, health scores, anomaly detection, and session diffs. No upload, no cloud, MIT. Ships a companion SKILL.md for in-agent audit runs.
      • beta - gated, so it never increases your bill. -31% input / -74% output tokens. Also an MCP server, CLI, and library (Rust, Python, Ruby, Swift, Kotlin, JS/WASM). MPL-2.0.
      • beta - platform) — Fork-and-rebrand multi-tenant web app for running Hermes agents: auth, chat, files, fleet management (Supabase; needs an Agent37 Cloud key)

  • Where Do I Start?

    • wondelai/skills - platform skills library that works with Hermes and other agents. Or pick up [litprog-skill](https://github.com/tlehman/litprog-skill) (75+ stars) if you want literate programming support across Hermes, Claude Code, and OpenCode.
    • hermes-workspace - native workspace with chat, terminal, and a skills manager. If you need multi-agent fleet management, [mission-control](https://github.com/builderz-labs/mission-control) (3.7k+ stars) handles task dispatch and cost tracking at scale.
Programming Languages
Python 37 TypeScript 20 Shell 6 JavaScript 6 Rust 3 HTML 2 Go 1 Dockerfile 1 Swift 1
Categories
Skills & Plugins 73 Tools & Utilities 28 Integrations & Bridges 27 Domain Applications 14 Related Projects 5 Multi-Agent & Swarms 4 Forks & Derivatives 4 Official Resources 4 Guides & Documentation 3 Where Do I Start? 2 Detection & Media Forensics 1 License 1 Operational Playbooks 1 Related Lists 1
Sub Categories
Deployment 62 Skill Registries & Discovery 27 agentskills.io Ecosystem 25 Plugins 22 Community Skills 21
Keywords
hermes-agent 27 mcp 21 ai-agents 19 claude-code 16 openclaw 15 llm 12 nous-research 11 hermes 10 ai-agent 10 agent-skills 9 agents 8 mcp-server 8 ai 7 self-hosted 7 claude 7 autonomous-agents 7 claude-skills 7 hermes-plugin 6 automation 6 model-context-protocol 6 typescript 6 developer-tools 6 skills 5 anthropic 5 codex 5 agent-tools 5 plugin 5 memory 5 ai-memory 5 claude-code-plugin 4 chatgpt 4 skill-md 4 cursor 4 python 4 agentic-ai 4 llm-agent 3 llm-agents 3 ai-skills 3 prompt-engineering 3 atropos 3 local-first 3 llm-tools 3 clawdbot 3 moltbot 3 openai 3 open-source 3 agent-skill 3 llmops 3 agent-orchestration 2 workflow 2