45 specialized judges that evaluate AI-generated code for security, cost, and quality.

Coming soon.

CLI wrapper for the Judges code review toolkit.

Playwright-based coding-agent tests, comparisons, explicit judges, and viewer

Evaluate and improve AI agents from runs, traces, judges, and feedback. Compare candidates, cluster failures, measure lift, and gate releases.

Autonomous AI engineer — observes, judges, builds, ships

Ohbem judges your Pokemon GO IVs.

Data-driven hooks: stored conditions that, when an LLM classifier judges them matched, invoke a tool. Cross-runtime (node + browser).

Coding-agent evaluation and improvement with deterministic assertions, optional LLM judges, and first-class human-in-the-loop runs.

Execution motor for Harmony playbook stages. Runs exactly one stage per invocation: worktree, role-separated subagents, held oracle, gate evidence. It never routes, never judges, never pushes.

Project-agnostic visual AI tester: captures what an app actually renders, judges it against UI best practices and per-project rules via the local Claude Code CLI, verifies acceptance criteria, and tracks findings in an adjudicated backlog.

Witness recording for agent runs — harness hooks emit a journal the agent cannot decline to write, and `nullius witness validate` judges it.

Makes the Pi agent follow your project's rules. Jev judges every write against your pi-warden.md and quotes the broken rule back to the agent, names slop, breaks stuck loops, calls out unverified done claims, compresses large tool output, and holds the ra

LLM-judge service for the @unotest ecosystem: judges free-form text (chat replies, generated content) against a natural-language rubric and returns a structured pass/fail verdict with reasoning. Runs as a small HTTP service (`npx @unotest/judge`) or in-pr

Contract-synthesis + match-judge engine for ggui's handshake. Synthesizes or repairs a conforming DataContract from an agent's draft, judges blueprint-match candidates for reuse, and validates contract structure + novelty — the primitives composed by deci

Judges a change before it lands: a secret-scan attempt (attested, or actually run via the ./secret-scan subpath's verified gitleaks acquisition), a change's task record, its review evidence, and policy drift, each reported as satisfied/violated/indetermin

Evaluation SDK for AgentV - build custom code judges

Tells you when your Pi session has reached a good point to compact: a standalone extension that judges completed checkpoints instead of leaving you to guess.

Your design system dies one pull request at a time. This makes sure it doesn't. A guard that judges only the lines a change adds, against the system the repo already has, and names the on-system value the author probably meant.

Self-learning habit engine for the DeepSeek Harness — detects user-correction signals, judges habits with a low-cost model on threshold, settles candidates behind a two-level human gate