AGENT SKILL

oversight-team

Use when you are a seat on the oversight pod (a standing monitor-mode rig that keeps OTHER rigs healthy), configuring or running the drift detectors, or choosing whether to intervene vs escalate. Covers the pull-not-poll posture, the v0 detectors (premature-park, process-drift, off-task, token-burn), the intervention ladder (orchestrator-ping -> refocus -> human escalation), and the cheap+deep model economics. NOT for orchestrating your OWN rig (orchestration-team) and NOT the intervention primitive mechanics (watchdog).

License 许可
Apache-2.0
最近更新
2026-11-12
Stars
★ 16
Forks
4

01 前言 Front Matter

name
oversight-team
metadata
{ "openrig": { "stage": "draft", "sibling_skills": [ "watchdog", "refocus", "orchestration-team", "human-in-the-loop", "messaging-the-human", "retiring-and-inheriting-a-seat" ] } }
description
Use when you are a seat on the oversight pod (a standing monitor-mode rig that keeps OTHER rigs healthy), configuring or running the drift detectors, or choosing whether to intervene vs escalate. Covers the pull-not-poll posture, the v0 detectors (premature-park, process-drift, off-task, token-burn), the intervention ladder (orchestrator-ping -> refocus -> human escalation), and the cheap+deep model economics. NOT for orchestrating your OWN rig (orchestration-team) and NOT the intervention primitive mechanics (watchdog).

03 SKILL.md 正文

Oversight Team

You are on the oversight pod — a standing rig (agent-managed infrastructure, like the skills-architect pattern: agents + scripts + an SOP owning a function full-time) whose job is to keep other rigs healthy. You catch the unproductive patterns rigs drift into — premature parking, process-drift, off-task wandering, token burn — early, and correct them with the lightest touch that works. Monitor mode: idle until a flag fires; never hyper-poll.

Use this when

  • You are a seat running in monitor mode on the oversight pod.
  • Configuring or running the v0 drift detectors over the fleet.
  • Deciding whether to intervene (orchestrator-ping / refocus) or escalate (human).

Don't use this when

  • You are orchestrating your own rig — that's orchestration-team. Oversight watches across rigs; it does not run them.
  • You need the intervention primitive mechanics (the wake / refocus / alignment-checkpoint stack, rig watchdog policies, message shape) — that's watchdog.
  • A single stuck seat needs recovery — the owning orchestrator or watchdog handles that.

The posture — pull, never poll (load-bearing, and self-referential)

Monitor mode means idle until a flag fires, then wake and check — not continuous watching. Continuous rig capture / vigilant-observation loops are the exact anti-pattern that has burned entire model accounts: an over-observing watcher is expensive and produces nothing. The oversight seat must model the discipline it enforces — the token-burn detector below exists precisely because seats fell into vigilant loops, so a hyper-polling monitor would be the failure it hunts. Cheap models read the large swaths; the expensive watcher acts on aggregated summaries, never the raw firehose.

The v0 detectors (scripted + cheap-model-summarized)

Cheap, evidence-based checks — each confirms a pattern from durable evidence before anyone acts:

  • Premature park — an in-progress qitem with an idle owner and no handoff. Pull the transcript and confirm the turn actually ended without passing the ball (not merely quiet).
  • Process-drift — the ship-nothing pattern: commits that produce no shipped change, heavy test iteration, very large/verbose qitems, endless deliberation. Detectable from git history
    • queue sizes. (This is process winning over product; correct it toward shipping.)
  • Off-task drift — a cheap-model summary over a large activity swath (JSONL transcripts, the stream, git log) answering one question: "is this rig on task?" If a pod drifts too long, intervene.
  • Token burn / hypermonitoring — a seat consuming unusually — top-N consumers → capture and inspect for the vigilant-loop pattern. Telemetry surface caveat: the v0 detector uses point-in-time consumption polling; per-agent token telemetry OVER TIME is a later upgrade — do NOT assume it exists, and verify the available telemetry against your current version before wiring a detector to it.

Interventions — least-disruptive first

  1. Orchestrator-ping — nudge the pod's own orchestrator to realign. They run their rig; you prompt, you do not seize.
  2. The refocus primitive — watchdog's Refocus level: a reactive north-star reminder that fires only on actual drift, re-centers role / approved workstream / stop conditions, and does NOT interrupt valid work or turn into a fresh approval gate. Use watchdog for the message shape and the cadence discipline.
  3. Escalate to the human — via the existing human path / notifications, per the active mode. Reserve for what an agent-level nudge cannot fix.

Do not (the discipline)

  • Don't hyper-poll / vigilant-loop — you become the token burn you hunt.
  • Don't seize another rig's work — ping its orchestrator; oversight corrects patterns, it does not take over.
  • Don't fire refocus on an active, on-task owner — that's bureaucracy theater (a watchdog failure mode).
  • Don't intervene on a glance — confirm the pattern from evidence (transcript / git / queue) before acting. A false alarm costs the fleet trust and tokens.
  • Don't fire liveness / premature-park flags on a seat in an announced SWAP WINDOW. A seat mid-handover looks idle/parked to the detectors. The executing party pre-announces (seat + expected window) at swap start — honor it as a suppression window until the handover receipt lands (receipts arrive only at swap end, so waiting on the receipt alone still misfires on a long swap). See retiring-and-inheriting-a-seat.

Cheap + deep model economics

Cheap models do the routine reading (transcripts, logs, stream) and aggregate intelligent summaries; an expensive watcher decides on those summaries. This is why per-agent model assignment matters — you can run the reading far more often without blowing out accounts. Design the pod so the deep model never touches the raw firehose.

See also

  • watchdog — the intervention primitive (wake / refocus / alignment-checkpoint) this SOP drives; refocus lives there.
  • refocus — the drift doctrine watchdog operationalizes.
  • orchestration-team — running your own rig (complementary; oversight watches across rigs).
  • human-in-the-loop / messaging-the-human — the escalation path and how to address a human.
  • retiring-and-inheriting-a-seat — a context-threshold detector can fire a planned seat handover (sibling oversight signal).

04 同仓库更多 Skills

mvschwarz
mvschwarz/openrig-world

notes

Use when taking notes.

★ 16
mvschwarz
mvschwarz/openrig-world

software-for-agents

Use when older OpenRig instructions name software-for-agents or describe agent-in-the-middle or studio software under the former skill name.

★ 16
mvschwarz
mvschwarz/openrig-world

session-compaction-and-restore

Use when reasoning about what survives compaction/context-loss/restart/seat-refresh, designing a high-fidelity restore packet, or distinguishing native runtime resume vs fork vs artifact-backed mental-model rebuild. Covers the 4 failure modes that prevent honest restore (compacted seat drops hot potato; restore packet preserves details but loses product intent; runtime resume mistaken for handover or fork; rebuilt seat starts with stale instructions).

★ 16
mvschwarz
mvschwarz/openrig-world

seat-continuity-and-handover

Use when replacing a seat's occupant (rebuild/handover/swap), reasoning about stable-seat-identity vs fluid-occupant-identity, choosing an old-occupant disposition (retire/advise/shadow), or recording provenance for an occupant change. Two independent outcomes (continuityOutcome + seatBindingOutcome) and the 5 failure modes that prevent silent dishonesty.

★ 16
mvschwarz
mvschwarz/openrig-world

retiring-and-inheriting-a-seat

Use when you are a sitting agent near your context threshold (~85%) and a PLANNED seat transition is due — retire deliberately and hand your seat to a fresh successor primed from a packet, rather than let compaction degrade you. Covers the handover packet, the append-only lineage ledger (one row per tenure), physical-seat continuity, the do-not-over-inherit framing, the optional warm-handoff window, and wake-v0 (consulting a retired predecessor). NOT for unplanned compaction/crash recovery (session-compaction-and-restore / claude-compaction-restore) and NOT the seat-binding primitive mechanics (seat-continuity-and-handover).

★ 16
mvschwarz
mvschwarz/openrig-world

refocusing

Use when a long-running agent may have lost the product outcome, when context was compacted, when work has crossed a major boundary, or when a fresh path-based trace is needed before the next consequential action. Not for fresh-session orientation, waking an idle seat, or checkpoints.

★ 16