Chaser Agent

Open SourceArchitect and Engineer

A governed, review-first source-intelligence agent harness extracted from ChaseOS principles.

Outcome

Source Card Harness V0 separates source claims, evidence, uncertainty, action candidates and memory candidates so a human reviews before anything is trusted.

Technology

  • Python
  • Deterministic review artifacts
  • Evidence modelling

Scope and boundaries

Not claiming a production autonomous agent, live provider routing, or a browser-control adapter.

Related build logs

← All projects

I take on a small number of projects at a time.

Available for selected agentic AI, automation, full-stack product and technical architecture work.