John Smith 3f1a35dcd2 docs: re-audit at 9e805ad - HR-6.1 + HR-5.5 confirmed fixed, live gate 22/0
Triad re-audit (Alpha/Beta/Gamma + cross-examination) finds zero open
findings: using-superpowers vendored/bound, tab -p deny materialized.
verify-rigs.sh 22/0, audit_loop self-test passed, regen idempotent.
Sentinel stays FINDINGS_DETECTED pending networked live checks.
2026-10-19 20:26:52 -04:00
2026-10-19 02:31:42 -04:00
2026-10-19 02:31:42 -04:00
2026-10-19 02:31:42 -04:00
2026-10-18 19:16:31 -04:00
2026-10-18 23:41:48 -04:00
2026-10-19 02:31:42 -04:00

Heretek-Rigs — 51-Agent OpenRig + Pi + OpenCode Go Fleet

Rigs (8)

  • executive-and-architecture
  • fullstack-and-web-studio
  • systems-and-mobile-foundry
  • game-dev-and-sim-rig
  • cloud-sre-and-data-platform
  • qa-and-verification-gauntlet
  • defensive-sec-and-red-team
  • heretek-omni-fleet (3-seat command + 15 pods + 51 members + rig.all-51.yaml)

Model

  • opencode-go/muse-spark-1.3-contributor:high via https://opencode.ai/inference/go/v1.

Secret provisioning

  • $HERETEK_SECRETS/opencode-go.key with mode 600.
  • auth.json uses !cat (never inline secrets).

Makefile targets

Target Purpose
generate Deterministically generate AgentSpecs, RigBundles, Roster, Registry
verify Self-testing verification gate
test Unit tests for seat rendering and permissions
bootstrap-seats Provision Pi seats across all rigs
seed Seed one rig before launch
gate Gate-test one role against mock OpenAI
guard-profiles Fail-closed sidecar profile guard
sidecars-up Start sidecar containers
sidecars-down Stop sidecars, preserve volumes
bundle-check rig bundle check across all 8 rigs
audit-loop Adversarial audit and remediation loop

Sidecars

  • Sidecar profiles note: enable only needed profiles via SIDECAR_PROFILES.
  • guard-profiles: mobile/devsec isolation — never co-schedule --profile mobile with devsec or web-research.

Seat seeding

  • Seed before rig up / expand / add.
Languages
Python 78.7%
Shell 11.9%
JavaScript 3.7%
YARA 2.6%
CodeQL 1.4%
Other 1.6%