Explore
MCP servers and skills: search, filter by where they run, and compare.
SWE-benchmarks-gen-harnessKentoShimizu
HostedGeneration harness for SWE-bench Verified experiments
SkillLink only★ 2Few stars
MCP servers and skills: search, filter by where they run, and compare.
Generation harness for SWE-bench Verified experiments
SkillLink only★ 2Few stars
SWE-bench repository patch localizer and regression verifier (Devin SWE style)
SkillLink only★ 8
Is this SWE-bench-style task fair to grade an agent on? Agentic screening with verified evidence, measured against human annotations. Also an agent skill: /fairtask
SkillMITFew starsNo key
Reproducible recon/craft/audit agent pipeline for SWE-bench Verified. Official-graded, codex-attested, GPL-3.0. Run it yourself.
SkillGPL-3.0★ 8No key
Personal SWE-bench harvested from your own dev sessions — replay real fixes across agent harnesses and models, score against your own verified patch
SkillMIT★ 1Few starsNo key
Best-of-N for coding agents with a same-model verifier, run your task N times in isolated git worktrees and apply the verified winner. A /ultra command for opencode, plus a standalone CLI.
SkillMIT★ 4Few starsNo key
Placebo-controlled benchmark of the most-starred coding agent skills
SkillMIT★ 19No key
SkillApache-2.0★ 4Few starsNo key
Find the agent-memory benchmark that fits your system in one offline command.
SkillMITFew starsNo key
Autonomous multi-agent AI coding assistant coordinating 21 specialized sub-agents with SWE-bench RAG
SkillLink only★ 1Few stars
Agentic SDLC orchestration for Claude Code — transcribe, triage, plan, implement, review, ship
SkillMIT★ 2Few starsNo key
Token governance for Claude Code: the top model directs, execution goes to the cheapest adequate means. Routing kernel, hook-enforced pre-budget, telemetry, statusline with plan quotas.
SkillMIT★ 4Few starsNo key
Claude Code plugin: plan on your main model, code on Sonnet xhigh, review on Opus. The expensive model never writes app code.
SkillMIT★ 1Few starsNo key
Claude Code skill: Claude Opus conducts and judges, GPT-6.1-Sol executes via the Codex CLI. Supervision every 30 min, council-backed decisions, HTML reports, Codex quota statusline.
SkillMIT★ 3Few starsNo key
Spec-to-task lifecycle manager for AI coding agents. Decomposes specs into atomic, dependency-ordered tasks. Agents execute with 2 tool calls instead of hundreds.
SkillMIT★ 8No key
🧪 the Evaluation Operating System — OEC (Observe → Evaluate → Control) for foundation models, agents, and business objectives
SkillApache-2.0★ 1Few starsNo key
挂帅·军师——双模型协作操典：STRONG 模型运筹、CHEAP 模型冲锋。Dual-model collaboration doctrine (orchestrator + advisor), adapters for Claude Code / OpenCode / Codex.
SkillMIT★ 1Few starsNo key
Claude plans, Codex / OpenCode Go / DeepSeek execute: a quota-aware dispatcher that routes coding tasks to the cheapest capable agent, in isolated git worktrees with hard budgets.
SkillApache-2.0★ 2Few starsNo key
Portable agent skill: live LLM leaderboard data from artificialanalysis.ai — Intelligence, Coding & Agentic indices, speed, pricing. No API key, zero deps (Node.js stdlib).
SkillMIT★ 4Few starsNo key
Deterministic Multi-Agent TDD Harness & Verification FSM for Autonomous Software Engineering
SkillMIT★ 1Few starsNo key
Automated benchmark patch verification harness & fail-to-pass matrix (SWE-bench)
SkillLink only★ 8
Test-Driven AI Development: GraphRAG-powered test impact analysis that reduces AI code regressions by 72%
SkillMIT★ 2Few starsNo key
Score Claude Code skills with deterministic checks, safety gates, eval fixtures, and ship/revise/reject scorecards.
SkillMIT★ 1Few starsNo key
Who stands behind it
