What I'm working on right now

Shipping right now

Last 14 days of pushes to agent-airlock, agent-audit-kit, proofloop, mnemo, provael, and this site.

• agent-audit-kit16c4815

  2026-10-03
  ci: publish the container image for linux/arm64 too (#885)
• agent-audit-kit7c5c91c

  2026-10-03
  chore: refresh public coverage page
• provaelf1dbc53

  2026-10-03
  docs: four comments the code had outgrown (#317)
• agent-audit-kit5d57cd4

  2026-10-03
  Disposition #879 to #883 and cut 0.6.15 (#884)
• provael460b48a

  2026-10-03
  docs(claude): integration gates, attestation binding, five more fixes (#316)

Writing this month

• Substack2026-10-02
  My App Says It Is Not Therapy. Its Page Title Says Therapy.
• Medium2026-09-29
  Stop Treating LLM Monitors as Controls. Gate the Tool Call Instead
• Medium2026-09-25
  My Precision Badge Was Right About the Wrong Files
• Medium2026-09-23
  Before You Delete Your MCP Servers, Count Their Tools

Talking

Open to podcasts, conference talks, and 1:1 advisory sessions on agentic AI, LLMOps, and AgentOps.

Reading

• Project Glasswing announcement, Anthropic

  Claude Mythos is the upstream attacker model. Decides what agent-audit-kit needs to detect next.

• OWASP Agentic Top-10 RC

  Mapping current agent-audit-kit coverage against the upcoming RC release.

• Salt Security 1H 2026 report

  88% of enterprises had agent-security incidents: agent-airlock's market sizing.

• VLA red-teaming: SmolVLA, π0, OpenVLA papers + LIBERO

  Current focus: building Provael, the red-team & assurance layer for physical AI. Mapping which vision-language-action attacks transfer from simulation to real policies.