ruflo-workflows

GAIA benchmark dispatcher — run, submit, validate, and track leaderboard scores against the Princeton HAL benchmark

Report cumulative GAIA API spend and project cost for planned configurations

Show measured benchmark runs stored across sessions in the gaia-runs memory namespace

Fetch and display current HAL GAIA leaderboard scores and our positioning

Execute a GAIA benchmark run — shells out to gaia-bench run, streams progress, and writes JSON results

Specialized agent for executing GAIA benchmark runs, monitoring progress, and analyzing results

Specialized agent for packaging, signing, and coordinating HAL leaderboard submission of GAIA benchmark results

Workflow automation specialist for creating, executing, and managing multi-step processes

Side-by-side comparison of ruflo vs HAL vs other GAIA harnesses — capability gaps, design decisions, and improvement roadmap

Diagnose why a GAIA question failed — extract trace, classify failure mode, and propose a fix. Use when a GAIA benchmark run reports a failed/incorrect task_id and you need to root-cause it before resubmitting.

Walk through a complete GAIA benchmark→submit flow — from key resolution through HAL-compatible package generation

Author a workflow — either an MCP workflow template (persisted, lifecycle) or a native .claude/workflows/*.js orchestration script (agent/parallel/pipeline fan-out)

Run a workflow — drive an MCP workflow lifecycle (execute/pause/resume/cancel) or invoke + resume a native .claude/workflows/*.js orchestration via the Workflow tool