SMALL TOOLS · REAL INTERACTIONS
Ideas you cantouch.
Compact experiments that expose how I think about software structure, agent authority and AI retrieval. They run entirely in the browser and make no claim to be production analyzers.
01Repository intelligence
Architecture visualizer
SAMPLE / COMMERCE-API
Reveal a deterministic service map and inspect where the sample architecture asks for review.
DEPENDENCY MAP / 06 NODESAWAITING SCAN
02Agent authority
Approval-state simulator
Goal receivedChange preparedHuman approvalExecute
PROPOSED ACTION / INFRASTRUCTURE
Scale a production service
Increase the sample payment service from three to six replicas. This simulation never performs an external action.
03Retrieval inspection
RAG evidence viewer
QUERY / SAMPLE INCIDENT
“What changed before checkout latency increased?”
- 01SCORE 0.94
incident-4821.md
Payment latency began immediately after the cache policy change.
- 02SCORE 0.87
redis-runbook.md
Checkout sessions require an eviction policy that preserves TTL.
- 03SCORE 0.73
deploy-214.log
Configuration rollout completed two minutes before the first alert.