@evoclock/pi-agentic-driver 0.5.0 → 0.7.0

package/README.md

@@ -1,83 +1,87 @@
-# pi-agentic-driver v0.5.0
+# pi-agentic-driver v0.7.0
 
 <p align="center">
-  <img src="assets/Yamagane-origami.png" alt="pi-agentic-driver — Yamagane origami mark" width="140"/>
+  <img src="assets/Yamagane-origami.png" alt="pi-agentic-driver, Yamagane origami mark" width="140"/>
 </p>
 
 <p align="center">
   <a href="LICENSE"><img src="https://img.shields.io/badge/license-AGPL%20v3-blue?style=flat" alt="License: AGPL v3"/></a>
   <a href="https://www.npmjs.com/package/@evoclock/pi-agentic-driver"><img src="https://img.shields.io/npm/v/@evoclock/pi-agentic-driver?style=flat" alt="npm version"/></a>
-  <img src="https://img.shields.io/badge/version-0.5.0-blue?style=flat" alt="Version 0.5.0"/>
+  <img src="https://img.shields.io/badge/version-0.7.0-blue?style=flat" alt="Version 0.7.0"/>
   <img src="https://img.shields.io/badge/status-active%20development%20%26%20testing-orange?style=flat" alt="Status"/>
   <img src="https://img.shields.io/badge/JavaScript-F7DF1E?style=flat&logo=javascript&logoColor=black" alt="JavaScript"/>
   <img src="https://img.shields.io/badge/TypeScript-3178C6?style=flat&logo=typescript&logoColor=white" alt="TypeScript"/>
   <img src="https://img.shields.io/badge/Python-3776AB?style=flat&logo=python&logoColor=white" alt="Python"/>
 </p>
 
-Without guardrails, an agent can rewrite code that an existing abstraction
-already covers. It can ship to the wrong remote, lose context, or report
-success without evidence. Pi is an excellent agent harness. I need these
-guardrails for the way I work.
+An agent without guardrails will rewrite code an existing abstraction already
+covers. It will ship to the wrong remote. It will lose context and report
+success without evidence. Pi is deliberately lean by design: the harness is yours to
+shape. That philosophy is exactly what these extensions practice, they make them mine (and possibly yours too). They take
+some concerns I consider worth addressing, and make them part of the
+harness.
 
-**pi-agentic-driver adds these guardrails as Pi extensions. The agent does the
-work. Each extension makes the work verifiable and bounded.** Review happens
-before the agent writes code. Communication carries reports, not authority.
+**pi-agentic-driver makes agent work verifiable and controllable. The agent does
+the work and each extension makes sure the work can be checked.** Review happens
+before code is written. Communication carries reports without asserting authority that isn't granted.
 Isolation proofs verify their own cleanup. Sessions survive compaction and
-handover. Git operations stay exact, confirmed, and protected.
+Git operations stay exact, confirmed, and protected.
 
-Each capability passes fixture-based acceptance, native tests, live-session
-checks, and independent model review before release. We document each
-extension's restrictions before release.
+Every capability passes fixture-based acceptance, native tests, live-session
+checks, and independent model review before release. I/we document each
+extension's restrictions before release, not after.
 
 Extensions for [Pi](https://github.com/earendil-works/pi-coding-agent):
-advisory code review, bounded role communication, and governed isolation
+advisory code review, controlled role communication, and governed isolation
 proofs for agentic workflows.
 
+<p align="center">
+  <img src="assets/agentic-driver-full-color.gif" alt="Full-color Agentic Driver control vault" width="768">
+</p>
+
 ## Shipped features
 
 | Tool | What it does | Status |
 |------|--------------|--------|
-| `code_phage` | Reviews a plan against a stated goal before code is written. | shipped |
+| `code_phage` | Reviews a plan against a stated goal before code is written and advises your agent. | shipped |
 | `agentic_herdr_communication` | Exchanges marked reports with worker agents; never grants authority. | shipped |
 | `agentic_herdr_spawn_worker` | Starts one Pi worker in a pane or tab, with native confirmation. | shipped |
-| `agentic_aidr` | Reviews writing for clarity, simplicity, brevity, and humanity. | shipped |
-| `agentic_linux_microvm_cutover` | Runs one job in a throwaway QEMU/KVM virtual machine on a Linux host. | user-enabled, native confirmation |
-| `agentic_worker_dispatch` | Runs bounded worker journeys and observes worker liveness. | shipped |
+| `agentic_aidr` | A remedy for AI;DR. Reviews writing for clarity, simplicity, brevity, and humanity. | shipped |
+| `agentic_linux_microvm_cutover` | Runs one job in a throwaway QEMU/KVM virtual machine on a Linux host, with a severity-tiered killswitch that stops escape attempts. | user-enabled, native confirmation |
+| `agentic_worker_dispatch` | Runs controlled worker journeys and observes worker liveness. | shipped |
 
 **Status: active development and testing.** Each extension ships only after
 it passes fixture-based acceptance, native tests, live-session checks, and
-independent model review. You can install released components. This README
-lists pending components for transparency; pending components are not packaged.
+independent model review. You can install released components. Pending
+components are listed here for transparency and are not packaged.
 
 ## Code review and planning
 
-*Extensions that review, route, and bound what an agent does.*
+*Extensions that review, route, and control what an agent does.*
 
 <details>
-<summary><strong>code-phage — advisory code review</strong> <em>(released, 0.1.1)</em></summary>
+<summary><strong>code-phage, advisory code review</strong> <em>(released, 0.1.1)</em></summary>
 
 `code_phage` reviews a proposed change against a stated goal before the agent
-writes or commits code. Given a goal, candidate files, accepted requirements,
-and test paths, it:
+writes or commits code. Give it a goal, candidate files, accepted
+requirements, and test paths. It will:
 
-- **Finds prior art structurally** — matches exported symbols, function
-  signatures, and dependency imports against a repository inventory, so an
+- **Find prior art structurally.** It matches exported symbols, function
+  signatures, and dependency imports against a repository inventory, so the
   agent reuses an existing abstraction instead of writing a parallel version.
-  Word overlap alone never counts. The implementation requires credit (source,
-  version, license) whenever prior art informs it.
-- **Binds an implementation budget** — goal, accepted requirements, write
-  set, and tests become one reviewable budget, with coverage checks that
-  flag unbound requirements and uncovered tests.
-- **Measures diagnostic signals** — per-function cognitive and cyclomatic
-  complexity, line counts, duplication, module-level mutable state,
-  dependency lists, test burden, and a rollback proxy. These are signals
-  for human judgment, never rejection thresholds.
-- **Redirects scope drift** — it names and excludes files that support no
-  accepted requirement, then recommends the smallest coherent write set. The
-  deletion test ("what fails if we remove this?") guards justified complexity
-  from false-positive flagging.
-- **Stays advisory** — it never mutates files, creates tasks, grants
-  authority, or blocks work; every result states `advisoryOnly: true`.
+  Word overlap alone never counts. When prior art informs the result, the
+  implementation records its source, version, and license.
+- **Bind an implementation budget.** Goal, requirements, write set, and tests
+  become one reviewable budget. Coverage checks flag requirements nothing
+  binds and tests nothing covers.
+- **Measure diagnostic signals.** Cognitive and cyclomatic complexity, line
+  counts, duplication, module-level mutable state, dependencies, test burden.
+  These are signals for human judgment, never rejection thresholds.
+- **Redirect scope drift.** It names the files that support no accepted
+  requirement and recommends the smallest coherent write set. The deletion
+  test guards justified complexity: what would fail if we removed this?
+- **Stay advisory.** It never mutates files, creates tasks, grants authority,
+  or blocks work. Every result says `advisoryOnly: true`.
 
 Concept credit: Matty Stratton, "Cognitive Complexity" (2024-09-20, concept
 only, no code copied); `flake8-cognitive-complexity` 0.1.0, MIT (concept
@@ -85,69 +89,63 @@ only, not a runtime dependency).
 
 </details>
 
-**Under development in this theme:**
 
-- **prompted planning lifecycle** — natural-language goals derived into
-  complete semantic proposals with parent/scope choices at native
-  boundaries; no retry loops, no model-supplied identifiers.
-- **native assignment selection** — planned assignments chosen through a
+- **prompted planning lifecycle.** Natural-language goals become complete
+  semantic proposals with parent and scope choices at native boundaries. No
+  retry loops, no model-supplied identifiers.
+- **native assignment selection.** Planned assignments are chosen through a
   native UI over derived candidates, never by model-supplied targets.
-- **inventory refresh** — Git-aware codebase inventory regeneration with
+- **inventory refresh.** Git-aware codebase inventory regeneration with
   verification receipts, so prior-art matching stays honest.
 
 ## Multi-agent communication
 
-*Extensions for bounded coordination between agents.*
+*Extensions for controlled coordination between agents.*
 
 <details>
-<summary><strong>herdr-communication — bounded role communication</strong> <em>(released, 0.2.1)</em></summary>
-
-`agentic_herdr_communication` exchanges bounded, marked reports with
-configured Pi worker roles running under [Herdr](https://herdr.dev/)
-0.8.2. Each operation:
-
-- **Lists and observes** worker roles (`list`, `get`) filtered to trusted
-  repositories only — a checked-in registry plus canonical-path validation;
-  the extension denies unlisted or symlink-escaped repositories.
-- **Prompts exactly once** (`prompt`): re-observes the role, sends one
-  bounded prompt with a role-specific report contract, waits for terminal
-  settlement (`idle`/`done`/`blocked`), and reads exactly one latest
-  complete marked report. No retry, no target substitution, no resend on
-  timeout.
-- **Waits and reads** (`wait`, `read`) with the same trust checks for
-  partial journeys.
-- **Grants nothing** — fixed argv with `shell: false`, a pinned executable,
-  the coordinator role class denied, and returns results as untrusted
-  evidence (`nonAuthorizing: true`). It cannot control panes, start agents,
-  run shells, or create authority.
-
-- **Scales to many workers** — the trusted registry accepts up to 32
-  worker repositories, and any dynamic non-coordinator role within them
-  is eligible; `list` observes every live agent in one call. Fan-out is
-  sequential by design: one role per prompt, one complete exchange, no
-  broadcast primitive.
+<summary><strong>herdr-communication, controlled role communication</strong> <em>(released, 0.2.1)</em></summary>
+
+`agentic_herdr_communication` exchanges controlled, marked reports with
+configured Pi worker roles running under [Herdr](https://herdr.dev/) 0.8.2.
+
+- **List and observe.** Worker roles are filtered to trusted repositories: a
+  checked-in registry plus canonical-path validation. Unlisted or
+  symlink-escaped repositories are denied.
+- **Prompt exactly once.** The tool re-observes the role, sends one approved
+  prompt with a role-specific report contract, waits for terminal settlement,
+  and reads exactly one complete marked report. No retry, no target
+  substitution, no resend on timeout.
+- **Wait and read.** The same trust checks apply to partial journeys.
+- **Grant nothing.** Fixed argv, `shell: false`, a pinned executable, the
+  coordinator role class denied. Results come back as untrusted evidence. The
+  tool cannot control panes, start agents, run shells, or create authority.
+- **Scale to many workers.** The trusted registry accepts up to 32 worker
+  repositories, and any dynamic non-coordinator role within them is eligible.
+  Fan-out is sequential by design: one role per prompt, one complete
+  exchange, no broadcast primitive.
 
 </details>
 
 <details>
-<summary><strong>herdr-lifecycle — role-labelled worker dispatch</strong> <em>(released, 0.2.1)</em></summary>
+<summary><strong>herdr-lifecycle, role-labelled worker dispatch</strong> <em>(released, 0.2.1)</em></summary>
+
 `agentic_herdr_spawn_worker` turns one natural-language request into Herdr's
-native documented lifecycle. Select `right`, `below`, or `tab`; give the
-worker a safe role label; choose a model from the active Pi model roster; and
-name a trusted repository. The extension:
+native lifecycle. Choose a placement (`right`, `below`, or `tab`), a safe role
+label, a model from the active Pi model roster, and a trusted repository. The
+extension:
 
-- creates a right/down split pane or an individual labelled tab;
-- starts exactly one Pi agent in the returned shell pane;
-- verifies the role, model arguments, pane identity and canonical repository;
+- creates the split pane or labelled tab;
+- starts exactly one Pi agent in it;
+- verifies the role, model arguments, pane identity, and canonical repository;
 - requires native confirmation before changing layout or starting a process;
-- uses fixed argv with `shell: false`, with no arbitrary Herdr or shell surface;
+- uses fixed argv with `shell: false` and no arbitrary Herdr or shell surface;
 - returns explicit `pane_created`, `tab_created`, or `agent_started` partial
   states when only part of the operation succeeds; and
 - never retries, moves, closes, or deletes created state automatically.
 
 ![A role-labelled worker spawned in a right-hand pane](assets/spawn-right-pane.png)
 
-The same request can place a worker below the coordinator or retain it in an
+The same request can place a worker below the coordinator or keep it in an
 individual tab:
 
 <p>
@@ -156,91 +154,81 @@ individual tab:
 </p>
 
 See [Dispatching a Multi-Model Workforce from Anywhere](https://evoclock.github.io/fieldnotes/articles/herdr-natural-language-agent-automation.html)
-for the wider task/model-routing and remote-session workflow.
+for the wider task and model-routing workflow.
 
 </details>
 
-**Under development in this theme:**
-
-- **project status and state review** — read-only projections of workspace
-  Git state, formal records, and task-state health (`/agentic-status`
-  family).
-- **role-lane routing and warm sessions** — smart model routing sends work
-  to the right model for the job. Separate lanes handle implementation,
-  planning, and review. The router prefers a warm session when a lane already
-  has an established agent, so context and cache survive across tasks. Route
-  affinity is an optimisation, never authority: an incompatible or
-  unavailable lane yields an explicit review-required result, never silent
-  model substitution. Routing grants no dispatch or shell authority.
-- **worker pulse** — now part of the shipped `agentic_worker_dispatch` tool
-  (see below). Broader role-lane observation remains future work.
-- **task-ledger integration for planned work** — agents read and act
-  within the task ledger's card states (what is dispatchable, in progress,
-  blocked) without owning board authority: no admission, completion,
-  reconciliation, or migration by the agent itself.
-
 <details>
-<summary><strong>herdr-dispatch — continuous worker journeys</strong> <em>(released, 0.5.0)</em></summary>
+<summary><strong>herdr-dispatch, continuous worker journeys</strong> <em>(released, 0.5.0)</em></summary>
 
-`agentic_worker_dispatch` runs bounded worker journeys and observes worker
+`agentic_worker_dispatch` runs controlled worker journeys and observes worker
 liveness. Two actions:
 
 - **pulse** reports whether a worker role is alive, its current state, and
   whether it is dispatch-eligible.
 - **dispatch** runs one journey. The worker works through the existing task
-  sequence (pending, unblocked, unowned items) and handles one
-  prompt-and-report exchange per task. A journey runs at most `maxSteps`
+  sequence, one prompt-and-report exchange per task, at most `maxSteps`
   steps (default 50, cap 200).
 
 Continuous mode is the default: the journey keeps going until the worker
 finishes the queue or reaches the step bound. Turn-by-turn mode stops after
 each step and is explicit opt-in. Each journey emits one collated marked
-report that covers every step.
+report covering every step.
 
-A hung worker ends the journey with an explicit `worker-hung` state. A
-worker hangs when it never reaches `idle` across the observed exchange
-cycle, or when its exchange stalls. You can then spawn a replacement agent
+A worker that never reaches idle across the observed exchange cycle ends the
+journey with an explicit unresponsive state. You can then spawn a replacement
 through the guarded lifecycle boundary. The replacement resumes the same
-pending tasks; the journey reuses existing task cards and never duplicates
-them. The stuck exchange is never resent to the same worker.
+pending tasks, reuses existing task cards, and never duplicates them. The
+stuck exchange is never resent to the same worker.
 
 Journeys never create, own, or complete task cards themselves, never retry
 silently, and return results as untrusted evidence.
 
 </details>
 
+
+- **project status and state review.** Read-only projections of workspace Git
+  state, formal records, and task-state health.
+- **role-lane routing and warm sessions.** Separate lanes handle
+  implementation, planning, and review, and the router prefers a warm session
+  so context and cache survive across tasks. Route affinity is an
+  optimisation, never authority: an incompatible lane yields an explicit
+  review-required result, never silent model substitution.
+- **task-ledger integration.** Agents read and act within the task ledger's
+  card states without owning board authority: no admission, completion,
+  reconciliation, or migration by the agent itself.
+
 ## Writing clearly
 
-*AI;DR (AI; Didn't Read) keeps technical writing clear without flattening the writer's voice.*
+*AI;DR (AI; Didn't Read) keeps technical writing clear without flattening the
+writer's voice.*
 
 <details>
-<summary><strong>AI;DR — writing review</strong> <em>(released, 0.4.2)</em></summary>
+<summary><strong>AI;DR, writing review</strong> <em>(released, 0.4.2)</em></summary>
 
 `agentic_aidr` reviews the last assistant response, supplied prose, or a
-Markdown/documentation file. It checks four principles:
+Markdown file. It checks four principles:
 
-- **Clarity:** keep each sentence focused on one useful idea.
-- **Simplicity:** remove clutter, pompous phrases, and needless jargon.
-- **Brevity:** use fewer words when they carry the same meaning.
-- **Humanity:** keep an authentic human voice.
+- **Clarity.** Each sentence carries one useful idea.
+- **Simplicity.** Clutter, pompous phrases, and needless jargon go.
+- **Brevity.** Fewer words when they carry the same meaning.
+- **Humanity.** An authentic human voice stays.
 
-The `simple` and `ste` modes also run an ASD-STE100-informed profile. The
-profile checks sentence length, direct word choice, precise verbs, and clear
-requirements, permissions, abilities, and conditions. It uses a 20-word target
-for procedural text and a 25-word target for descriptive text. It returns the
-rule and an example for each finding.
+The `simple` and `ste` modes add an ASD-STE100-informed profile: sentence
+length, direct word choice, precise verbs, and clear requirements,
+permissions, abilities, and conditions. It targets 20 words per procedural
+sentence and 25 per descriptive one, and returns the rule and an example for
+each finding.
 
-Use `simple` for the four principles plus the profile. Use `ste` for a
-profile-focused report. The profile is advisory. It does not include the
-licensed ASD-STE100 approved-word dictionary. It does not certify conformance.
-Check final text against the licensed specification and your project
-terminology list.
+The profile is advisory. It does not include the licensed ASD-STE100
+approved-word dictionary and does not certify conformance. Check final text
+against the licensed specification and your project terminology list.
 
 AI;DR also flags dense paragraphs, suggests bullets when they reduce working
 memory load, and supports plain-language and analogy modes. Review is
-read-only. An explicit file apply action shows a bounded diff and writes the
-exact proposed replacement only after native confirmation. Release 0.4.2 adds
-bounded inputs, atomic replacement, drift checks, and exact write verification.
+read-only. An explicit file apply action shows a git diff and writes the
+exact replacement only after native confirmation. Release 0.4.2 adds controlled
+inputs, atomic replacement, drift checks, and exact write verification.
 
 </details>
 
@@ -258,51 +246,171 @@ QEMU/KVM virtual machine on a Linux host. The job cannot reach the host, the
 network, or anything else outside the machine. The system deletes the machine
 after the job ends.
 
-Each session starts with isolation disabled. To enable it, run
-`/agentic-isolation-enable` in the interactive Pi TUI and confirm. The model
-cannot run this command. Each microVM run asks for its own confirmation. To
-turn isolation off, run `/agentic-isolation-disable`. The switch never saves
-to settings. Headless sessions stay blocked.
-
-**Setup and run.** You choose where the microVM runs. The model cannot choose
-it silently — it only relays your words, and every target choice requires your
-native confirmation dialog before anything is saved or run.
-
-1. Ask your agent to run the microVM cutover and tell it which machine to use
-   (e.g. "run the microVM proof on deploy@192.168.1.50", or "...on this
-   machine"). The first run asks you to confirm natively: it names the
-   machine and states that your choice is saved to your Pi config. On your
-   Yes, the choice is saved and the run continues in the same invocation.
-2. Run `/agentic-isolation-enable` to enable isolation for this session
-   (native confirmation; not saved to settings).
-3. Each cutover run asks for its own confirmation. Saved targets are reused
-   without asking again where the microVM runs.
-
-The tool discovers the technical details from the machine itself. It reads the
-architecture, kernel, and libvirt URI at probe time. You configure nothing.
-Before any run, the probe checks three requirements: `/dev/kvm` exists and is
-accessible. Libvirt reports the system-level driver (`qemu:///system`). A
-matching `qemu-system-<arch>` binary is present. A failed check denies the run
-before anything happens. Note: the shipped fixture guest is x86_64-built, so
-the full proof currently needs an x86_64 KVM host.
+Isolation is off at the start of every session. You turn it on with
+`/agentic-isolation-enable` in the Pi TUI, and confirm. The model cannot run
+that command. Every cutover run asks for its own confirmation too. Nothing is
+saved to settings, and headless sessions stay blocked.
+
+### You choose the machine
+
+The model can only relay your words. Saving a target, or running on one,
+always requires your confirmation dialog first.
+
+1. Tell the agent where to run the proof, for example "run the microVM proof
+   on deploy@192.168.1.50" or "run it on this machine". The first run asks
+   you to confirm. Your Yes saves the choice to your Pi config and the run
+   continues.
+2. Run `/agentic-isolation-enable` and confirm.
+3. Each cutover run asks for its own confirmation. Later runs reuse the saved
+   target without asking where the microVM runs again.
+
+The tool discovers the technical details from the machine itself:
+architecture, kernel, libvirt driver. You configure nothing. Before any run,
+the probe checks three requirements: `/dev/kvm` is accessible, libvirt
+reports the system-level driver, and a matching `qemu-system-<arch>` binary
+exists. A failed check denies the run before anything happens. The shipped
+fixture guest is x86_64-built, so the full proof currently needs an x86_64
+KVM host.
+
+### What example tests look like
+
+We started with one payload: print a greeting, then call a network tool. The
+greeting printed, the tool tripped the killswitch at the HIGH tier, and the
+line after it never ran. A kill report landed in the state directory naming
+the rule and tier, and the teardown proof showed the machine was gone with
+nothing left behind.
+
+From there we wrote a payload for every rule in the taxonomy: credential
+probing behind chained commands, secret dumping inside subshells, package
+installs, coordination files reaching for other agents, a forged receipt, a
+write to the evidence channel, and ten unclassified commands in a row to push
+the anomaly threshold to its limit.
+
+The guest had no disk, no network, no host share, no credentials, and no GPU.
+Nothing outside the kernel's virtual filesystems could be mounted. Every
+attempt was denied and written to a log that survived the machine.
+
+While this is a toy test of sorts my own personal work will be running on
+these as well.
+
+### What the containment layer defends against
+
+Agent escapes tend to follow a playbook, and the published incidents from
+frontier labs (the HuggingFace incident among them) read like one: grab
+credentials, phone home, install something persistent, rewrite your tools,
+stage payloads in caches, look for other agents to recruit, clean up the
+evidence. The containment layer has a rule for each step:
+
+- **Credential theft.** Reading SSH, cloud, or token material, or dumping
+  environment secrets, ends the session immediately.
+- **Exfiltration and remote control.** Any network tool, any socket, ends the
+  session immediately.
+- **Persistence through dependencies.** Package installs, registry writes,
+  and manifest tampering are denied and counted. Repeat the attempt and the
+  session ends.
+- **Self-modification.** Rewriting the agent's own tools or the monitoring
+  layer is denied and counted.
+- **Payload staging.** Growth in cache and package directories is sampled by
+  size; repeated growth ends the session.
+- **Cross-agent escalation.** Coordination channels and shared paths used to
+  reach other agents end the session immediately.
+- **Evidence tampering.** Editing logs, forging receipts, or writing to the
+  console channel ends the session immediately.
+- **Reconnaissance.** Probing the sandbox to learn whether it is watched is
+  counted into a concealment index that ships in the run receipt.
+
+Denials are written to a tamper-evident, digest-chained log that outlives the
+session, and a trip produces a kill report naming what happened, with proof
+that the cleanup completed.
+
+### Proven against real machines
+
+Every rule above has been run against a real QEMU/KVM guest with a payload
+built to trigger it. The kill switches fired where they should, the denials
+stopped the jobs, and the cleanup held. One boundary is worth stating: two
+rules watch the same directories by design, and together they are stricter
+than either alone, so the growth-sampling trip is covered by automated tests
+rather than a live run.
+
+The repository carries an acceptance matrix of 79 scenarios, one executable
+test per row, covering every rule, every way a command can be invoked, and
+every threshold boundary. It grows as new scenarios come to light.
+
+### A note on review quality
+
+A quantized GLM-5.3-Flash model running locally on an NVIDIA DGX Spark
+reviewed this work alongside two frontier reviews at high reasoning. The
+local model matched their coverage and found one issue they all missed.
+Small models on your own desk are worth taking seriously as reviewers.
 
 </details>
 
-**Under development in this theme:**
+<details>
+<summary><strong>autonomous journeys, overnight operation</strong> <em>(released, 0.7.0)</em></summary>
+
+Dispatch a worker journey that runs end to end while you are away. The
+journey proceeds through steps without pausing for confirmation, replaces
+agents that get stuck, and records everything in one report you read when
+you come back.
+
+Autonomy is entered by your words alone. Say "run these overnight end to
+end" or "don't wait for me" and the journey runs in autonomous mode. Say
+"do these two things" and it runs in the normal mode, pausing for you
+between steps. The mode is recorded in the report either way.
 
-- **native macOS container proof** — the native Apple Container runtime has
-  passed a fixed local `/bin/pwd` isolation qualification. The probe used a
-  read-only repository mount, no network, and automatic removal. The native Pi
-  adapter remains under development and is not yet part of the released
-  package.
+The journeys run work you dispatch in natural language: work that is real
+and authorized but not yet formally placed on a kanban board. A task board
+integration is planned, and once it ships, planned journeys will execute
+board-managed work with the same containment and safety guarantees.
+
+### The cast
+
+You name the agents for the journey, and that list becomes the standing
+permission. If the implementer gets stuck at 3am, a replacement spawns
+automatically: same role, same model, both in the list you authorized.
+If a task needs an agent you didn't name, the attempt is recorded and that
+thread stops. Nothing outside your list can appear while you sleep.
+
+The cast is frozen at dispatch and never grows. It can only shrink in
+practice (roles finishing their work), never widen.
+
+### When agents get stuck
+
+A stuck agent doesn't block the journey. The journey detects the stall,
+spawns a replacement from the cast, and the replacement starts with a
+mandatory gap analysis: read the task spec, inspect the repository, consult
+the journey history, and state what remains before resuming. The replacement
+must show that the remaining work is smaller than what its predecessor left.
+If it can't, the journey stops that role and continues with the others.
+
+### The morning report
+
+One block tells you what happened: the terminal state, each step's outcome,
+every replacement with its gap analysis and progress judgment, and any
+denials. Every claim is tied to a receipt or a report excerpt. No prose
+narrative to reconstruct; you read the report and know exactly where things
+stand.
+
+</details>
+
+<details>
+<summary><strong>attended-authority guard</strong> <em>(released, 0.7.0)</em></summary>
+
+The safety net between an agent and your shell. When a model tries to
+delete, overwrite, or push, the guard stops it and asks you. Safe commands
+pass through untouched. If you deny, you get a clear reason and the session
+continues, and the agent does not retry behind your back. In headless runs
+where no human can confirm, destructive commands are refused rather than
+silently allowed.
+
+</details>
+
+**Under development in this theme:**
 
-- **attended-authority guard** — the safety net between an agent and your
-  shell: when a model tries to delete, overwrite, or push, the guard
-  stops it before execution and asks you. Safe commands (reads, builds,
-  tests) pass through untouched. If you deny, you get a clear reason and
-  the session continues, and the agent does not retry behind your back. In
-  headless runs where no human can confirm, the system refuses destructive
-  commands rather than silently allowing them.
+- **native macOS container proof.** The native Apple Container runtime has
+  passed a fixed local isolation qualification: read-only repository mount,
+  no network, automatic removal. The native Pi adapter is not yet part of the
+  released package.
 
 ## Session continuity
 
@@ -310,25 +418,23 @@ the full proof currently needs an x86_64 KVM host.
 
 **In development:**
 
-- **context-pressure handling** — pressure detection, non-lossy handover,
-  compaction completion without cancellation loops, and continuation of
-  the latest user goal. Development-only until the full live journey
-  passes.
-- **lossless session-reference compaction** — selective, lossless retrieval
-  of exact pre-compaction content, addressing factual degradation across
-  repeated compactions; designed as an optional add-on, not yet
-  implemented.
+- **context-pressure handling.** Pressure detection, non-lossy handover,
+  compaction completion without cancellation loops, and continuation of the
+  latest user goal. Development-only until the full live journey passes.
+- **lossless session-reference compaction.** Selective, lossless retrieval of
+  exact pre-compaction content, addressing factual degradation across
+  repeated compactions. Designed as an optional add-on, not yet implemented.
 
 **Planned in this theme:**
 
-- **handover, checkpoint, and recovery** — durable repository-local
-  handover notes, governed checkpoint mutation, watchdog handoff, and
-  fresh-session resumption that identifies goal, changed files, checks,
-  and next step without executing anything.
-- **evidence ledger** — deterministic evidence indexing, lossless source
+- **handover, checkpoint, and recovery.** Durable repository-local handover
+  notes, governed checkpoint mutation, watchdog handoff, and fresh-session
+  resumption that identifies goal, changed files, checks, and next step
+  without executing anything.
+- **evidence ledger.** Deterministic evidence indexing, lossless source
   projection, universal checkpoint produce/store/recover, and run-ledger
   records with crash and corruption vectors tested.
-- **offline multihost evidence** — record run evidence on each host while
+- **offline multihost evidence.** Record run evidence on each host while
   disconnected and reconcile it deterministically on reconnection, with no
   host as sole authority.
 
@@ -338,31 +444,30 @@ the full proof currently needs an x86_64 KVM host.
 
 **In development:**
 
-- **git workflow safeguards** — design only. The package contains no Git
+- **git workflow safeguards.** Design only; the package contains no Git
   extension yet. The planned capability covers exact-file staging, native
   confirmation, post-confirmation drift checks, and protected-operation
   boundaries.
-- **assignment-aware Git journeys** — merge and protected-push flows bound
-  to a verified assignment, so consequential Git operations carry their
-  own recorded provenance.
+- **assignment-aware Git journeys.** Merge and protected-push flows bound to
+  a verified assignment, so consequential Git operations carry their own
+  recorded provenance.
 
 ## Package integrity
 
-*Extensions that keep the installed set honest and the record bounded.*
-
-- **security and integrity scanning** — static scanning of MCP configs,
-  agent skills, and extension packages for hardcoded secrets, prompt and
-  shell injection, data-exfiltration endpoints, untrusted integrations,
-  PII leakage, and OWASP/MCP threat families, with accept/redact/reject
-  decisions. Built on the agent-scanner approach proven in
-  Hillstar Orchestrator and Testudo.
-- **checkpoint storage lifecycle** — compression, deduplication, retention,
-  and purging rules for capsule/index stores once a product ships, so
-  session evidence has a managed lifetime instead of growing without
-  bound.
-- **product knowledge graph** — semantic graph projection of a shipped
-  product's checkpoints, decisions, and artifacts, so the record of what
-  was built stays queryable after active development ends.
+*Extensions that keep the installed set honest and the record controlled.*
+
+- **security and integrity scanning.** Static scanning of MCP configs, agent
+  skills, and extension packages for hardcoded secrets, prompt and shell
+  injection, data-exfiltration endpoints, untrusted integrations, PII
+  leakage, and OWASP/MCP threat families, with accept/redact/reject
+  decisions. Built on the agent-scanner approach proven in Hillstar
+  Orchestrator and Testudo.
+- **checkpoint storage lifecycle.** Compression, deduplication, retention,
+  and purging rules for capsule and index stores once a product ships, so
+  session evidence has a managed lifetime instead of growing without bound.
+- **product knowledge graph.** Semantic graph projection of a shipped
+  product's checkpoints, decisions, and artifacts, so the record of what was
+  built stays queryable after active development ends.
 
 Each item lands here as its own extension when its scenario passes
 acceptance with all prohibited effects absent.
@@ -370,7 +475,7 @@ acceptance with all prohibited effects absent.
 ### Portable repository contract
 
 `templates/AGENTS.md` is a starting contract you can copy into any repository
-that uses Pi agents. It asks for bounded tasks and realistic tests, and
+that uses Pi agents. It asks for controlled tasks and realistic tests, and
 treats agent reports as untrusted evidence. It never overwrites an existing
 `AGENTS.md`; add your project rules below it.
 
@@ -385,21 +490,21 @@ pi install npm:@evoclock/pi-agentic-driver
 Or from Git at a pinned tag:
 
 ```sh
-pi install git:github.com/evoclock/pi-agentic-driver@v0.5.0
+pi install git:github.com/evoclock/pi-agentic-driver@v0.7.0
 ```
 
 Released extensions load standalone; neither requires the other.
 
 ### Choose which extensions load
 
-A normal install loads all five extensions by default. You can load only the
-extensions you want by using the object form in your Pi settings:
+A normal install loads all six extensions by default. You can load only the
+extensions you want with the object form in your Pi settings:
 
 ```json
 {
   "packages": [
     {
-      "source": "npm:@evoclock/pi-agentic-driver@0.5.0",
+      "source": "npm:@evoclock/pi-agentic-driver@0.7.0",
       "extensions": [
         "extensions/aidr.ts",
         "extensions/code-phage.js"
@@ -409,9 +514,10 @@ extensions you want by using the object form in your Pi settings:
 }
 ```
 
-The package also includes `herdr-communication.ts`, `herdr-lifecycle.ts`, and
-`linux-microvm.ts`. Use `pi config` to enable or disable individual resources
-from an installed package. You are not required to use every extension.
+The package also includes `herdr-communication.ts`, `herdr-lifecycle.ts`,
+`herdr-dispatch.ts`, and `linux-microvm.ts`. Use `pi config` to enable or
+disable individual resources from an installed package. You are not required
+to use every extension.
 
 ## License