Turn AI coding sessions into a local, searchable retro graph
Project overview
ax ingests local agent sessions, tool calls, plans, skills, Git history, and related signals into a typed graph. It lets you recall past work, inspect skill and tool usage, estimate costs, surface friction, and review proposed changes to project instructions. A session-end retro loop feeds accepted lessons into later work. Ingest is local by default; publishing profiles or sessions is an explicit opt-in.
Give AI agents code intelligence through an MCP language-server layer
Project overview
agent-lsp orchestrates existing language servers and exposes their code intelligence to AI agents through MCP. It keeps language servers warm between sessions, combines many routine calls into batch operations, and supports workflows such as caller analysis and speculative edits. The project documents 65 tools, 24 agent workflows, and CI-verified support for 30 languages.
General skillSoftware developmentModel Context Protocol
sk-requirements-engineering-theskill
@seekdevcore·Python
Bring requirements discovery, specification, and validation into AI development tools.
Project overview
Use the corpus as a Claude Code skill, managed plugin, or MCP server in clients such as Cursor, Cline, Continue, and Zed. It guides work from elicitation and hierarchical backlogs through user stories, BDD, acceptance criteria, estimation, validation, and traceability, with Canadian English and Brazilian Portuguese editions. The project is intended for defining and checking requirements, not for implementing application code.
General skillKnowledge managementModel Context Protocol
second-brain
@stancsz·Python
Give AI agents local, searchable memory backed by Markdown and SQLite.
Project overview
second-brain stores notes, relationships, and distilled long-term memories as readable Markdown, with a rebuildable SQLite cache for full-text and graph queries. Use it as a Claude Code skill for capture and proactive recall, or run its zero-dependency MCP server so other clients can read and write the same data. It requires Python 3.8 or later, supports Git-based sync, and is intended for personal-scale knowledge rather than massive document collections or team wikis.
MCP serverAI and agent developmentModel Context Protocol
skyll
@assafelovic·Python
Let agents discover and load relevant skills at runtime.
Project overview
Skyll exposes a REST API and MCP server that searches multiple skill sources, ranks matches, and returns complete SKILL.md content with optional references. Agents can choose and load instructions during a task instead of relying on preinstalled skills. A hosted service is available, or teams can self-host the server for more control.
MCP serverSecurity and privacyModel Context Protocol
open-skills
@instavm·Python
Run Claude skills locally in an isolated container from any MCP-compatible assistant.
Project overview
OpenSkills exposes Claude-style skills through an MCP server and executes them in an isolated local container, allowing assistants such as Claude Desktop, Gemini CLI, or Qwen CLI to process documents and images without cloud uploads. Path mapping lets existing skills run without modification and returns files to a local outputs directory. It currently requires an Apple Silicon Mac, macOS, and Python 3.10 or newer.
General skillMedia, audio and videoModel Context Protocol
video-reader
@MaximRomanMd·Python
Turns video and audio into frames, transcripts, metadata, and sound maps Claude can read.
Project overview
video-reader drives ffmpeg and ffprobe to extract frames, metadata, spectrograms, and waveforms from local media, with optional Whisper transcription and yt-dlp URL downloads. It exposes both standalone Python scripts and a Claude Code skill, while the core features require only ffmpeg. Frames are downscaled and capped by default so visual inspection does not overwhelm the context.
Share remembered context across AI agents, email, meetings, Slack, and CRM tools.
Project overview
Nex turns agent conversations and connected business sources into a shared knowledge graph, so facts recorded in one tool can be recalled from another without repeated explanations. Its setup command detects supported coding platforms and installs hooks, MCP integration, and slash commands, while a hosted MCP endpoint supports other clients. Using the service requires a Nex account and any desired data integrations.
Resolve scholarly identifiers into citations and bibliography exports
Project overview
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.
Audit AI skills for security and quality risks in CI or agent workflows.
Project overview
TAR Engine's audit layer discovers and checks SKILL.md files, Codex skill definitions, Claude commands, OpenCode configs, and adjacent helper scripts from the CLI, CI, or an MCP-compatible agent. Static rules run by default; semantic and adversarial analysis requires an explicitly supplied model key. The hosted mode sends submitted skill text to tarai.dev, so sensitive material can use the self-hosted backend instead.
Research Hungarian resale prices and draft listings for the right platforms.
Project overview
Magyar-eladó researches public Hungarian marketplace listings, estimates quick-sale, market, and patient prices, recommends suitable platforms, and writes platform-specific listings with photo folders and a posting checklist. Optional browser assistance stops before final submission by default; Facebook is never DOM-automated, while an explicit auto-submit flag is available at the user's own risk.
CLI appDesign and creative workModel Context Protocol
Muser
@connerkward·Python
Index and search local image folders by natural language through a CLI, UI, or MCP app.
Project overview
Muser indexes image folders on-device and supports text-to-image, image-to-image, color search, deduplication, provenance filters, clustering, and training-set export. You can use its CLI, local browser UI, or MCP app, while the core works offline without API keys or telemetry. Optional captioning and generation use external services and may cost money, so the project suits asset curation as well as local search.
General skillSecurity and privacyModel Context Protocol
sign-safe-skill
@lrafasouza·TypeScript
Decodes Solana transactions offline and gates them as SIGN, HOLD, or REJECT.
Project overview
sign-safe decodes an opaque Solana transaction before signing, classifies dangerous instruction shapes from the signer's perspective, and emits a SIGN, HOLD, or REJECT verdict plus machine-readable JSON. Its core requires no key, RPC, or broadcast; optional simulation and RPC enrichment are separate, and SIGN only means the transaction passed the current static policy.
MCP serverAI and agent developmentModel Context Protocol
agent-terminal
@jasonkneen·JavaScript
Control interactive command-line applications from AI agents without a visible terminal.
Project overview
agent-terminal lets an AI agent launch interactive CLI applications, capture their screen as ASCII text, and send keyboard input without a display. It can be used as an npm library or exposed through an MCP server for CLI testing, monitoring, and build automation. Built-in safeguards include command allowlists, workspace path validation, environment sanitization, input checks, and session limits.
General skillKnowledge managementModel Context Protocol
searching-apple-notes
@yinanli1917-cloud·Python
Search and write Apple Notes locally in Chinese or English
Project overview
searching-apple-notes lets Claude query and update an Apple Notes archive through natural language in Chinese or English. BGE-M3 embeddings and ChromaDB run locally on the Mac, with citations in answers and formatted notes written back; no API key, MCP setup, or third-party server is required, aside from downloading the model once.
General skillCommunication and collaborationModel Context Protocol
pith-skill
@VjAlbert·Python
Compresses long inter-agent payloads while protecting meaning.
Project overview
PITH compresses long natural-language handoffs between agents with local word-information scoring, while quarantining code, JSON, URLs, and paths. A negation check protects polarity, and a bounded sentence-length gate can reduce the pruning ratio and retry. Payloads below 10,000 characters pass through unchanged; the skill, MCP server, and Python CLI work offline without external APIs.
CLI appResearch and knowledgeModel Context Protocol
evidentia
@kgraph57·TypeScript
Verify medical citations and flag mismatched or fabricated identifiers before publication.
Project overview
Evidentia extracts DOI, PMID, arXiv, and NCT identifiers from medical writing and checks them against services including CrossRef, PubMed, OpenAlex, arXiv, and ClinicalTrials.gov. The CLI and MCP server classify citations as verified, bibliographically mismatched, hallucinated, or needing content review; the Claude Code skill adds a 15-criterion appraisal of how evidence is used. It is for writers, editors, and researchers, not clinical decision support.
MCP serverAI and agent developmentModel Context Protocol
Better-OpenCodeMCP
@ajhcs·TypeScript
Run OpenCode models from Claude through MCP with asynchronous task control.
Project overview
Better OpenCode MCP exposes the OpenCode CLI to Claude Code and other MCP clients, enabling selectable models and non-blocking long-running tasks. Calls return a task ID that can be monitored or supplied with additional input, while process pooling, timeouts, and persisted task state help manage concurrency and recovery. It requires Node.js and a working OpenCode installation.