Pinned Loading

1. llm-behavior-eval llm-behavior-eval Public

   Behavioural and red-team evaluation harness for chat assistants. Two-tier grounding checks: free deterministic regex first, LLM judge only where it earns the cost. Runs against a local model, OpenA…

   Python

2. support-triage-agent support-triage-agent Public

   Turns a support inbox into review-ready drafts. Retrieval-grounded, human-in-the-loop, no send path.

   Python

3. prd-figma-test-generator prd-figma-test-generator Public

   Turns a PRD (and optionally the Figma frames for the same feature) into a structured test suite: CSV for your test management tool plus a reviewable checklist. Reads the specification, not the code.

   Python

4. postgres-analyzer postgres-analyzer Public

   React + Express tool that reports PostgreSQL performance problems: slow queries, blocking locks, index health, bloat. Queries verified against a real server.

   JavaScript

5. api-contract-pii-monitor api-contract-pii-monitor Public

   Finds personal data in an OpenAPI spec: parameters, request bodies, responses, following $ref and composition. No database needed to analyse a spec.

   Python

6. postgres-dba-agent postgres-dba-agent Public

   Watches a PostgreSQL instance and, when something looks wrong, has an LLM write the tuning advice and post it to Slack as a readable card. Ships the Prometheus, postgres_exporter and Grafana stack …

   Python