MCP servers and skills: search, filter by where they run, and compare.
An MCP (Model Context Protocol) server for automated cell type annotation in scRNA-seq analysis using CellTypist with natural language.
PythonMITFew starsNo key
HeavySkill: Heavy Thinking as the Inner Skill in Agentic Harness
SkillMIT★ 2Few starsNo key
A meta-agent for generating multi-agent systems for a given task
SkillMIT★ 1Few starsNo key
Skill Devs within methodology WoT - wrangling of thought
SkillMIT★ 2Few starsNo key
Multi-persona AI review skill for Claude Code. Spawns 4-6 expert subagents in parallel + an adversarial skeptic, then synthesizes with a different model to avoid self-preference bias. Preserves dissen
SkillMITFew starsNo key
Multi-model consensus debate via the filesystem. LLMs propose, peer-review, rebut, vote and synthesize a group-confirmed answer. CLI + MCP.
PythonMIT★ 13No key
Multi-model workflow plugin for Claude Code: MAGI weighted voting + cross-CLI orchestration
SkillMIT★ 3Few starsNo key
Agent skills for team workflows: MR/PR review and testing (standard / hard / ultra)
SkillMITFew starsNo key
A council of seven cheap-tier model voters that resolves an ambiguous decision into a verdict — a harness-agnostic agent skill.
SkillMIT★ 1Few starsNo key
Model comparison pipeline — chains delegate_to_model, create_expert, execute_expert, consensus_vote across all 5 voting strategies
SkillMITFew starsNo key
Open-source AI council formations for multi-agent debate. 15 structured methodologies, three-valued voting, and sprint contracts.
SkillMITFew starsNo key
Evidence-based expert council for Claude Code: parallel cited web research first, then a large panel of independent personas, adversarial objectors, and a clear synthesis.
SkillMITFew starsNo key
Who stands behind it
