RyanAlberts

RyanAlberts/best-of-Agent-Harnesses

Cursor rule Cursor

Git author identity for commits so GitHub attributes contributions correctly.

RyanAlberts/best-of-Agent-Harnesses

Agent Claude Code

Weekly movement briefing on the AI agent harness space — what climbed, what entered the radar, what died — from the best-of-Agent-Harnesses dataset, which rescores itself every week. Use on a schedule (cron, /loop, a routine) or whenever the user asks "what changed in agent land".

RyanAlberts/best-of-Agent-Harnesses

Agent Claude Code

Picks the right AI agent harness for a described task or project. Use when the user asks "what harness/framework/agent tool should I use", compares agent frameworks, or starts an agent project without a stack decision. Grounded in the live best-of-Agent-Harnesses dataset, never in training-data memory.

RyanAlberts/best-of-Agent-Harnesses

Agent Claude Code

Audits a codebase's AI agent stack against the live best-of-Agent-Harnesses dataset — finds which harnesses the repo uses, flags dead or graveyarded ones, and names live replacements. Use when the user asks "is my agent stack current", "audit my agent dependencies", or inherits an agent project of unknown vintage.

RyanAlberts/best-of-Agent-Harnesses

MCP server Claude CodeCodexCursor +2

MCP server for best-of-Agent-Harnesses: opinionated harness recommendations (recommend/pickharness), search, and head-to-head decision guides over a hand-curated, weekly-rescored list of agent harnesses. Runs locally from the agent-harnesses-mcp Python package.

RyanAlberts/best-of-Agent-Harnesses

Hook Claude Code

One of 2 in settings.json

Runs before the agent uses a tool for Bash tool calls, executing guard.sh. From RyanAlberts/best-of-Agent-Harnesses.

RyanAlberts/best-of-Agent-Harnesses

Settings file Claude Code

One of 2 in settings.json

Agent settings declaring 1 hook event (PreToolUse) and 4 allowed tools.

RyanAlberts/best-of-Agent-Harnesses

Plugin Claude Code

Bundles 10 skills · 1,534 tokens together

Ten skills that test your agent harness on your own machine: sandbox reach, guardrails, 'tests pass' claims, rules the agent breaks, token waste, regressions after updates, MCP tool design, and a test drive of coding agents on tasks from your own git history.