Evaluation & Benchmarks

Search, vet and explain TESS transit candidates with provenance. Candidates, never confirmations.

Find RL environment graders that pay wrong answers: probes reward code, returns reproducible leads

Read local agent-run JSONL traces through MCP for failures, approvals, and run comparisons.

Guardrails for LLM answers: exact checks against sources, PASS/REVIEW/FAIL, signed records.

Authenticated, organization-scoped MCP tools for EvalGate quality evidence and planning.

Compare text LLMs across OpenRouter, Bedrock, Vertex AI and Foundry with automated grading.

Capture local agent traces and inspect failures, approvals, and run comparisons.

Verifies token/cost savings claimed by AI context-reduction proxies via MCP tools.

Wraps the slop-eval CLI as a single generic MCP tool for genericness scoring of AI UI output.

Independent, reproducible benchmark harness for agent-memory backends.

Run local prompt regression checks from prompt-response logs in AI coding assistants.

Verify a number before an agent asserts it — evals, backtests (Deflated Sharpe), signed receipts.

Wraps the fpp CLI as a single generic MCP tool for freelancer client payment-risk checks.

PDF-to-Markdown extraction that audits its own output and flags any extractor's silent drops.

MCP over HTTP; every tool is an x402-paid endpoint settled per call in USDC on Base or Solana — no accounts, no API keys.

MCP server that lets an agent run an independent Verigent verification of its own capability (proof-or-zero) and read the result. Free first run.

Measures the cost, tokens and duration of a completed agent workflow from its session log.

AWS security scanner with attack chain detection, IAM privilege escalation, and fixes

Profile analysis, hashtag research, content calendars, competitor benchmarks, viral scoring

Check generated audio, video and stills for defects. Returns a verdict, not more media.