Essay 6.2b — The Markov Phasic Brain, Part 3 of 13.

Essay 6.2 drew the full transition map and named the tool-restriction discipline that makes each phase distinct — the rules of the cycle. Before we open each compartment one at a time, here is the whole set at a glance. Each line below names the phase, its essence, and what it is on the hook to produce — a quick tour to hold in your head before the per-phase deep-dives begin, starting with OBSERVE in the next essay.

The Phases at a Glance

IDLE — the meta-state between cycles. Lifecycle management only — a narrow allowlist opens, and everything else stays shut. ⓘ - Unlock the job-management CLI — the lifecycle surface (show, focused, list, update, activate, focus, pause, complete, approve). Creation and graph mutations live elsewhere. - Unlock the phase CLI (advance, current, cycle, exit-gmode) — agent-callable advance only goes idle → observe - Keep the always-on infrastructure running: memory-file edits, plus a small named-script allowlist on the IDLE Bash gate — interaction_summary/scripts/summary.sh for cross-conversation summaries, brain_guard/scripts/self-compact.sh for context-window compaction, and plugin_integrity/scripts/lock-cmd.sh for the universal active-lock close-out. Every other shell command exits blocked. ⓘ - Block reads, project edits, CLAUDE.md edits, web access, general shell - Job creation happens automatically: top-level via prompt-handler.sh (the user’s prompt itself creates the job when none is focused); dependent jobs via CONDENSE step 3 consuming [PENDING-JOB] markers. The agent does not call job.sh create itself.

OBSERVE — gather context before any plan can form. ⓘ - On cycle 1, expand the job’s objective from its short creation stub into a full statement of intent — OBSERVE’s distinctive cycle-1 work; the Stage decision (the set-plan-file call PLAN must make) comes next, in PLAN - Populate the working-memory CLAUDE.md files with relevant context - Dispatch parallel research subagents and synthesize their returns - Refuse code edits — the only allowed write target is CLAUDE.md - Cross the exit threshold only after enough investigation has happened

PLAN — turn observations into a binding contract. ⓘ - Record the Stage decision in cycle 1 via an explicit set-plan-file call — false for a single-cycle job, a .md/.yaml name for a multi-cycle one; PLAN itself never writes the plan file — EXECUTE creates it - Declare the altered list — the set of dirs whose CLAUDE.md the agent edited during OBSERVE or PLAN; EXECUTE will be allowed to write project files inside each of those dirs exactly (no ancestor or nested dirs) - Write acceptance criteria VERIFY will check against - Refuse code edits — the contract is what gets written, not the work

EXECUTE — build what the plan declared, in checkpoints. ⓘ - Edit project files, but only inside the altered list — the merged set of CLAUDE.md files OBSERVE and PLAN together declared, frozen at execute entry - Materialize every artifact the seed agent produces — code, the .md plan in cycle 1 of a Stage-2 job, the .yaml plan in cycle 1 of a Stage-3 job, anything else with a path; EXECUTE is the universal file-creator - Favor small, focused checkpoint commits over one long uncommitted run; the intermediate-commit mode keeps checkpoints cheap, so the pattern costs nothing - Capture execution notes in the working CLAUDE.md so the cycle stays narratable - Delegate file work to execute subagents (sequential by default, a small in-flight ceiling); keep the main session on the spine

VERIFY — judge prior work with independent eyes. ⓘ - Run scripts and validators; refuse all code edits in this phase - Dispatch auditor subagents to read the executed work without bias - Write pass/fail results into CLAUDE.md and the plan file - Refine the focused job’s plan file (.md or .yaml) — but editing it blocks forward-advance and forces a backward step to PLAN; the plan is refined in place and persists, nothing is approved or sealed - Hand completion to CONDENSE — VERIFY has no job-completion authority; [JOB-COMPLETE] is a CONDENSE-only ceremony, a phase away - Review the focused job’s open dependencies and, when the audit reveals one is no longer needed, either unlink it but keep the work (job.sh remove-dependency — the child survives as standalone work) or unlink and abandon it (job.sh void-dependency — the child is marked voided, a terminal-abandon state) — the lifecycle-symmetry partner of CONDENSE’s add-dependency - Route the cycle forward to CONDENSE, or backward to whichever prior phase the failure points at

CONDENSE — consolidate the cycle’s learnings into the brain. ⓘ - Walk a strict staged waterfall that routes content to its durable home - Consume marker types the prior phases dropped into footers - Compress the working CLAUDE.md back to a clean state - Own all job creation (create, create-dependent) and dependency additions (add-dependency) — the phase with the right cycle-wide context for graph mutations - Lock forward to idle; no escape hatch back to verify

GMODE — the freestyle side-channel from any phase. ⓘ - Enter via a [GMODE] user question with a substantive reason (currently a roughly 100-word floor in the prototype) - Run unconstrained work — no OPEVC tool-restriction guards apply - Exit explicitly with a clean git tree; the home phase resumes atomically - Host work that doesn’t fit the OPEVC ceremony — deadlock fixes, plugin maintenance, custom workflows

That is the operational shape — and every edge, every locked tool, every gated boundary exists for one payoff: the agent can no longer skip the thinking a phase was built to force. The rest of this essay series opens each compartment one at a time to show exactly how that discipline does its work.

What you would customize

The discipline is the architecture; the specific edges and the specific allow-lists are this prototype’s calibration. Several surfaces are open to the next architect.

The architect would tune the backward map. The prototype lets verify roll back to several destinations — execute, plan, or observe — picked by the failure shape. A seed that runs short, surgical cycles might collapse the menu to one (always observe; let the next cycle re-plan from scratch). A seed running long structural sweeps might widen the menu to include condense for re-routing learnings without re-executing. ⓘ

The architect would tune the gmode usage policy. Gmode is the freestyle escape hatch; the prototype reserved it for plugin maintenance because the prototype was building the seed agent itself. A user-facing seed might keep almost all work inside OPEVC and never enter gmode. A research-heavy seed might push routine literature scans through gmode rather than inflating every cycle’s observe phase. ⓘ

The architect would tune the per-phase tool allow-lists. The current cuts — read-only in observe and plan, scripts-only in verify, full-write-inside-scope in execute, brain-only in condense — encode this prototype’s notion of cognitive separation. A seed wanting a stricter observe could ban the web entirely. A seed wanting a looser verify could allow targeted code edits inside named directories. The guards are code; the cuts are decisions. ⓘ

The architect would tune the metacog menu — how much reflection each phase owes before it may advance. The exit gate asks for a minimum number of reflection ops per phase, and that minimum is a dial, not a constant. A seed running fast, low-stakes work might require a single reflection op at each boundary. A seed doing careful, high-consequence work might demand several — a first-principles pass in OBSERVE, a second-order-effects pass in PLAN, an inversion check before building. A consulting practice could require a premortem lens before any client deliverable enters its build phase — the gate ensures the plan was stress-tested before a single hour is billed. The reflection lenses themselves come in two tiers: a small core that every phase must run, plus richer mental-model lenses the architect can switch on phase by phase. The knob decides how much of the agent’s compute goes to examining the work versus doing it. ⓘ

What the architect would not customize is the rule that each phase publishes its allow-list and the guard enforces it against every tool call. The principle is the floor: a phase whose restrictions are advisory is not a phase. ⓘ

The shape lifts off the prototype into work that has nothing to do with seed agents. A patent attorney shaping a prior-art-review seed could compartment the work into pull-references, extract-claim, cross-check, and draft-opinion phases — each with its own tool fence, each with its own write scope — and the transition map would refuse a draft-opinion → pull-references slide that smuggles unverified prior art into the brief.

The discipline here is friction, not mathematical enforcement: every guard depends on the agent reading and obeying the injected voice, and the slow-downs (the rhythm gates, the commit shapes, the substantive-reason floor) buy the operator the time to intervene before a bypass admits. Gmode is the documented escape hatch — the seed agent can route around the OPEVC ceremony when the work genuinely needs it, and the cost is the long-form justification that surfaces the bypass to the operator.

You now have the whole map in your head: each state with its essence and its output, and the customization knobs that let a different architect re-cut them. The rest of the series opens each compartment one at a time. The next essay opens the first one — OBSERVE, the read-wide-write-once entry phase that decides what the rest of the cycle will work on.

Essay 6.2b — The Markov Phasic Brain, Part 3 of 13.

Previous: Essay 6.2 — The Discipline and the Map — the full transition map and the tool-restriction discipline that makes each phase distinct. Next: Essay 6.3 — OBSERVE — Read Wide, Write Once — the entry phase, its wide sources, and the read-before-write rhythm.