The orchestrator Pi is retired (2026-05-31). All agents now run on the
cc-ci-orchestrator VM (NixOS, loops user, /srv/cc-ci). The VM is a
direct tailnet peer to cc-ci — no SOCKS proxy, no userspace tailscaled,
no ProxyCommand. Updated across all affected files:
AGENTS.md
- Remove Pi from reboot description; migration complete (not "parked")
- cc-ci access: direct ssh, not via proxy
kickoff.md
- Prerequisites: direct tailnet peer, not proxy
- Host deps: NixOS (not apt)
- Fallback/Incus: b1 reachable directly, no --proxy curl flag
plan.md §1 + §1.5
- §1 bootstrap: direct SSH, check tailscale status (not restart proxy)
- §1.5 intro: "VM" not "sandbox host"; no proxy
- Credentials table: remove TS_AUTH_KEY row; update cc-ci SSH row
- Replace "Tailscale connection (proxy)" subsection with direct-peer description
plan-orchestrator-migration.md
- Mark COMPLETE (2026-05-31); historical record only
plan-phase1c-full-reproducibility.md
- Incus access: direct, not via SOCKS proxy
prompts/builder.md + prompts/adversary.md
- cc-ci access language only: direct ssh, no proxy restart instructions
- adversary: *.ci.commoninternet.net via plain curl, no proxy flag
REBOOTS.md
- Retitle for VM; note Pi retired; Pi entries marked historical
systemd/cc-ci-loops.service
- User/Group/HOME/PATH: notplants → loops
- Remove cc-ci-tailscaled.service dependency (no proxy on VM)
- Add note about nix/configuration.nix as the authoritative VM declaration
test-e2e-testme-acceptance.md
- tailscale status: no --socket flag
- ssh to throwaway: no ProxyCommand
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The builder wedged at the context limit (garbled output) — alive but matching
none of heal_session's signals (dead/FATAL/limit), so the watchdog left it
stuck. Fix: loops now declare every wait, and the watchdog reboots a wait that
never resumes.
- plan.md §7 + both prompts: cap every wait at 10 min (chunk longer waits);
before going idle, the loop's FINAL line must be `WAITING-UNTIL: <ISO8601 UTC>`
(the resume time, matching its ScheduleWakeup); run /compact proactively at
~80% context to avoid wedging near the limit.
- launch.sh: new stall_check (runs every 30s signal tick) — reboots a loop idle
>= STALL_IDLE (300s) when it has NO current WAITING-UNTIL marker as its last
message OR is past the time the marker named; a healthy paced wait (marker
present, before its time) is left alone. Complements heal_session's
dead/FATAL/limit cases. Reboot is safe — loops re-orient from git + STATUS.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Cheap guard against the deploy/git divergence: a fix built locally but uncommitted/un-pushed is a
guaranteed Adversary cold-build mismatch. Added to the builder prompt claim discipline + plan.md
§6.1. (Lighter than binding the deploy to a git rev — iteration speed + the Adversary's
cold-from-git verify is the real safety net.)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replaces the brittle markdown prose-match ("Gate: … CLAIMED, awaiting Adversary") with detection of
the loops' conventional commit prefixes on origin/main: a new `claim(...)` commit pings the
Adversary; a new `review(...)` commit pings the Builder. Edge-triggered on the origin/main SHA
(append-only — no force-push), no file parsing, can't mis-route. The loops already use these
prefixes consistently; codified as a load-bearing contract in plan.md §6.1 + both prompts so it
stays reliable. INBOX detection unchanged (pushed-state, file-routed).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
launch.sh watchdog now runs a fast (~30s) handoff_check alongside the heavy (300s) restart/DONE
check: when the Builder writes a CLAIMED gate it pings the Adversary to verify now; when the
Adversary updates REVIEW.md it pings the Builder to proceed (edge-triggered, reads local clones).
So a pending handoff resolves in <~30s instead of a whole idle interval. Pacing revised: the
Adversary may idle freely when nothing's pending (no pointless re-verify/busy-poll) and is woken
by the watchdog; Builder waits on the ping + a fallback ~2-4m self-poll. kickoff documents the
new "handoff signalling" role.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Root cause of "both waiting": parked-at-gate was lumped into the long idle sleep, so a
pending handoff sat while both loops slept on desynced timers. Fix: three cases — (1) in
flight → ~4m; (2) BLOCKED ON THE OTHER LOOP (Builder at CLAIMED gate / Adversary awaiting a
fix) → ~4m poll for the counterpart, never long-idle; (3) genuinely nothing pending → ~10-15m.
Adversary: a CLAIMED gate is immediate top-priority; otherwise run background probes, rarely
idle while Builder is active. Builder: keep an unblocked item in hand to rarely be fully gated.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
§7 pacing + Builder/Adversary prompts: idle/parked sleep lowered to ~10–15 min so the next
unit of work (or a gate claim) is picked up without long gaps. Unchanged: ~4m polling while
a build/deploy is in flight; keep polling something clearly in-flight rather than treating
it as idle; don't spin on a minutes-long build. Adversary aligned to §7 for consistency.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Planning + launch + setup material for the cc-ci Co-op Cloud recipe CI server:
plan.md (single source of truth), kickoff/launch supervision, and the
Builder/Adversary loop prompts. Secrets (.testenv) and runtime dirs are gitignored.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
