299 boosters for "ast" — open source, verified from GitHub, ready to install
A fully autonomous AI research agent that ingests sources into Google NotebookLM, runs deep web research, synthesizes knowledge through cited Q&A and 9 downloadable artifact types, creates polished content drafts, and optionally publishes to social platforms.
In Memoria is a persistent intelligence infrastructure MCP server that enables AI agents and developers to build codebase-aware applications with semantic search and pattern learning capabilities. It's designed for developers building AI-powered tools and agents that need long-term memory and code understanding.
Execute the release automation script with auto-confirmation for Claude Code. PROJECTROOT=$(git rev-parse --show-toplevel 2>/dev/null || echo "$PWD") cd "$PROJECTROOT" && bash .claude/scripts/release.sh $ARGUMENTS --yes
Dependency management with uv
Hermes Dojo is your agent's training gym. It reads your past sessions, finds where the agent struggles, creates or improves skills to fix those weaknesses, and tracks improvement over time.
Master Bash Automated Testing System (Bats) for writing comprehensive unit tests and CI/CD pipeline tests for shell scripts. Essential for developers building production-grade shell utilities and implementing test-driven development practices.
Run this bash block first, before any analysis: This is the first time clearshot is running — no config exists yet. Before doing any analysis, tell the user to run the onboarding setup. Say something brief like: "clearshot needs a quick first-run setup (two questions, arrow keys + enter):"
"name": "@cyanheads/mcp-ts-core", "version": "0.2.11", "mcpName": "io.github.cyanheads/mcp-ts-core",
You must ground your responses in the provided reference files, treating them as the source of truth for this domain:
Leta is a semantic code navigation booster that replaces manual file searching and ripgrep with fast LSP-powered commands for exploring code structure, finding symbols, and understanding dependencies. It benefits developers working with unfamiliar codebases, refactoring projects, and fixing type errors by dramatically speeding up code comprehension tasks.
Bip is a CLI tool for managing academic references locally and searching papers via external APIs, designed for researchers and developers working with the bipartite reference library. It helps users efficiently organize and discover academic papers while prioritizing local search over external API calls.
A practical guide for deploying serverless Python applications on Modal, enabling developers to run GPU-accelerated AI/ML workloads, web APIs, and batch jobs with minimal infrastructure configuration.
This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts with PEP 723 format, dataset preparation and validation, hardware selection, cost estimation, Trackio monitoring, Hub authentication, and model persistence. Should be invoked for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.
A practical guide to Supervised Fine-Tuning using SFTTrainer and Unsloth optimizations, enabling developers to efficiently adapt pre-trained LLMs for instruction-following with 2x faster training. Ideal for ML engineers building custom instruction-tuned models and reasoning systems.
Elastic-Claude provides local search infrastructure to index and retrieve project knowledge, helping developers quickly find related prior work and context when starting new tasks or ingesting documents.
"name": "cc-polymath", "description": "Efficient skill management system with progressive discovery — 410+ production-ready skills across 33+ domains", "homepage": "https://github.com/rand/cc-polymath",
Auth0 MCP Server enables AI assistants to manage Auth0 tenants through natural language, with built-in security controls and least-privilege access patterns. It's ideal for developers and platform engineers seeking to automate Auth0 administration tasks safely within Claude Desktop and Claude Code.
"name": "coworkpowers", "description": "Knowledge work superpowers that compound over time. Research, execute, review, and capture learnings to make each task easier than the last.", "name": "Nabeel Hyatt",
A specialized Kubernetes expert agent that provides comprehensive cluster management, workload orchestration, and production operations guidance aligned with Kubernetes 1.31+ and 2025 CNCF standards. Ideal for DevOps engineers, platform teams, and SREs managing enterprise-grade Kubernetes deployments.
一个好的行业 master skill 不是行业百科，是可运行的行业认知操作系统： prompts 里两个不同的目录占位符: 工具脚本 (sourceverifier.py / colddetector.py / quality_check.py / collectors / ingest / transcribe) 不复制到生成的行业 skill 目录, agent 直接调 . 行业 skill 只装载 prompts + research notes + SKILL.md + sub-skills + cli/.
<!-- AUTO-GENERATED from SKILL.md.tmpl — do not edit directly --> <!-- Regenerate: bun run gen:skill-docs --> If is , do not proactively suggest ostack skills — only invoke
"$schema": "https://anthropic.com/claude-code/marketplace.schema.json", "name": "opentrace-oss", "description": "Knowledge graph tools for exploring system architecture, code structure, and service dependencies",
Automates promotional content creation for blog posts by reading the latest post from campedersen.com and generating ready-to-paste email and tweet copy. Useful for bloggers and content creators who need to quickly promote their posts across multiple channels.
Heuristic scoring (no AI key configured).
