mathiasandClaude Opus 5
294d9e2fc5
The atlas described stage 05 as agentsquad only, while headless single-issue execution has run through mathias/dispatch for months and gained substantial machinery in the 2026-08-17→19 session. Four nodes added, and the stage now names both paths rather than one. - dispatch · headless issue runs — the 1-minute poll over allow-listed repos, and the rule that CI-green on the pushed commit is the only done-signal. - Two independent executors — claude or kimi (berget/Kimi-K2.6 via LiteLLM, EU-metered), routed per issue. Marked as a win, and as risk-bearing: it exists because a single provider's empty account stopped all CAD for six weeks. - Fail-closed tool policy — one policy shared by both executors, writes confined to the checkout (symlink-resolved after an escape was found live), per-run turn and USD ceilings. - Read-only PR review — status:review, no write tools, and the grounding rule that a review citing no changed file is dropped rather than published. Deliberately NOT added: the risk-tiered agentic review with ntfy escalation. That is designed (dispatch#48) and not built, and the atlas is a map of what runs, not of what is planned. Every node carries the plain-language layer the build test enforces. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011PL5XcG68SvQqz3vGVbWbf
cad-atlas
A visual atlas of the Continuous Agentic Development (CAD) workflow — from a captured signal to a deployed k3s pod.
One screen, reel-style: Signals → TELOS → Strategic session → Spec/Oath → Human gate → agentsquad → CI → CD → pod, looping back to TELOS. It renders the CAD audit chain, which doubles as the regulated-industry audit artifact.
The CAD audit chain is the visualization data. Phase C renders it once and serves two masters: observability + compliance.
Phases
- A — static hero viz (current): self-contained
internal/web/static/cad-atlas.html, served at/. SVG spine, animated pulse, dashed feedback bus, replay/slow-mo. - B — generated-from-source: render from brain docs + workflows + infra manifests (no drift).
- C — live trace viewer: hydrate from the
assessor-loopledger +session_log+ Gitea run API + Flux events.
Quickstart
task check # lint + vet + test (CI gate)
task run # → http://localhost:8080
Context
Full project context, the 9-stage workflow, the three governance gates, the dogfooding model, and all brain deep-links + external references live in .context/PROJECT.md. Any clean agent session should read that first (brain_get <path> for the linked notes).
Governance (dogfooded)
This repo is built through the workflow it depicts: dispatch-allow-enabled, build increments governed by a var-go Oath in their spec issues. Bootstrapping honesty: the Oath is defined but cmd/vargo-gate is not yet wired into this repo's CI — advisory until it is. See PROJECT.md.