Rework the Development Plan around live testing and the revised architecture

SSH execution and the agent as a container image come first (Phase 1), then the new loop as a thin end-to-end slice, widened milestone by milestone with a live run after each. Outline, Vikunja and per-agent VMs are gone; the vault, the infrastructure repository and the agent host replace them. The survey and issue map are refreshed. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01PPQ14L3LikyefCMFBqEbKE
2026-10-02 19:41:48 +00:00
1 parent b7f1f3371c
commit 7b5f0dfd18
@@ -1,314 +1,241 @@
 # Nisse — Development Plan
-**Status: proposed for discussion.** How to get from the code as it is to the full
+**Status: agreed (PR #84); reworked in PR #92 after the architecture revision (PR #90).** How
-architecture in `docs/Architecture.md`, in what order, and how each step is
+to get from the code as it is to the architecture in `docs/Architecture.md`, in
-accepted. The architecture says *what and why*; this document says *how and when*,
+what order, and how each step is accepted. The architecture says *what and why*;
-and is revised as work lands.
+this document says *how and when*, and is revised as work lands.
-It follows the format the architecture asks agents to use (D2): phases of
+**The priority is live testing.** Work is ordered so the agent runs live against
-milestones, each with a goal, scope and acceptance criteria, and each marked
+real resources as early and as often as possible, and what each run finds steers
-**pre-approved** or **needs discussion**.
+what comes next.
 ---
 ## 1. Where we are
-A survey of the code on `main` (2026-09-30) against each architecture section.
+A survey of `main` (2026-10-02) against the architecture. **Done** means it exists
-**Done** means it exists and matches the architecture; **partial** means something
+and matches; **partial**, something exists but needs changing; **none**, not
-exists but needs changing; **none** means not started.
+started.
-Only `nisse-dev` has real code (about 10,000 lines in `nisse-dev/src/nisse_dev/`).
+Only `nisse-dev` has real code; `nisse-core` and `nisse-ops` are placeholders.
 `nisse-core` and `nisse-ops` are placeholder packages.
 | Section | Status | Notes |
 |---|---|---|
-| **P** Principles | — | Guidance; each milestone below applies them |
+| **L1** Work model | partial | Work items table with an active pointer (M1.1); no issue intake, queue, parking or blocked items |
-| **L1** Work model | partial | One work item at a time with a minted ID; but the state store is a single row (`agent_state`, `CHECK (id = 1)`), not a work-items table with an active pointer. No duties, parking or switching (only `nisse-ops` needs duties) |
+| **L2** Agent loop | partial | `control.py`, a fixed state machine with no-stranding, `stop`, `abandon`. Replaced by the new loop (Phase 2) |
-| **L2** Agent loop | partial | `control.py`: a fixed state machine (arch → milestone gates → executing → PR open), with the no-stranding rule, `stop`, `abandon`, restart questions. To be replaced by the conversational loop |
+| **L3** Authority | partial | Revision binding and arrival stamping (`gates.py`); scope is still plan approval, not the issue and its conversation |
-| **L3** Authority | partial | Revision binding and arrival stamping are done (`gates.py`, #74). Approvals are gates, not permissions; no pre-approved milestones, no trust dial |
+| **L4** Classifier | none | |
-| **L4** Action classifier | none | |
+| **L5** Conversation | partial | Matrix transport, mentions, lenient commands; no threads, no question handling, little explanation |
-| **L5** Conversation | partial | Matrix transport, mention handling, case-insensitive commands. Only the planner can ask (`ask_human`); no question handling, no explanations or progress prose, a misleading reply in `IDLE` |
+| **L6** Context | partial | Context fitting, per-role views and orchestrator facts (M1.2); no scope or agent memory |
-| **L6** Context and memory | partial | Context fitting is done (`context.py`). The transcript isn't scoped per work item and has no attribution or orchestrator facts (#78). No scope or agent memory |
+| **L7** Canonical log | partial | Work item and actor on every event, `fact` (M1.1, M1.2); no `classifier_verdict`, `knowledge_retrieved` |
-| **L7** Canonical log | partial | `canonical.py`: user, assistant, tool call/result, plan, gate, delegation, compaction events. Missing: work item and actor on every event, `fact`, `classifier_verdict`, `knowledge_retrieved` |
+| **L8** Durability | mostly done | WAL store, intents, boot reconciliation, PR adoption, comment redelivery; to move onto the new loop |
-| **L8** Durability and recovery | mostly done | WAL store, intents, boot reconciliation, adopting a published PR (#75), comment cursors and redelivery (#68, #76), Quint model (`specs/`). Needs moving onto the new loop and the work-items schema |
+| **L9** Models and roles | partial | LiteLLM, delegation, `ask_expert`, eval capture; roles are still planner/executor, no capability profiles |
-| **L9** Models, roles, helpers | partial | LiteLLM client and catalog, context budgets, delegation with passthrough and caps, `ask_expert`, tool-choice handling, eval capture. Roles are executor/planner/delegate/consultant with a frontier planner; no `driver`, no capability profiles, no driver-chosen escalation |
+| **L10** Knowledge vault | none | |
-| **L10** Knowledge store | none | |
+| **L11** Injection | partial | Operator allowlists on every channel; no classifier |
-| **L11** Prompt injection | partial | Operator allowlists on every channel; no classifier, no structured intake beyond Gitea's JSON |
+| **L12** Secrets, logging | partial | Structured state-change log; secrets in `.env`; no per-operation handout or redaction |
-| **L12** Secrets, identity, logging | partial | Structured state-change log (`oplog.py`); secrets from `.env` files; no Vaultwarden, no LLDAP identity |
+| **L13** Execution | none | Tools run locally in the agent's process. The token leaks into the clone (#89) |
-| **D2** Documents as workflow | none | Plans are posted to a Gitea issue; no project `ARCHITECTURE.md` or build plans in the repo |
+| **D1–D2** Workflow | none | Plans posted to a Gitea issue and gated; no issue intake, labels, or documents in the repo |
-| **D3** Workspace | partial | Cloned once and reused as-is; no fetch, no fresh workspace, no preflight (#62, #27) |
+| **D3–D4** Workspace, git | partial | Idempotent publishing; no fresh workspace (#62), `git add -A` (#80) |
-| **D4** Git and PRs | partial | Idempotent publishing (#75), committed work published (#63); `git add -A` sweeps in artifacts (#80); branches cut at publish from whatever `HEAD` is (#62) |
+| **D5** Review | partial | PR conversation comments only (#69), watch stall (#70) |
-| **D5** Review and merge | partial | Review loop on PR conversation comments; inline comments unseen (#69); watch stall (#70); operator merges; no trust dial (#65) |
+| **D7–D8** Deployment, projects | none | Runs from a venv as a systemd unit on this box; one project in config |
-| **D7** Deployment | none | Bring-up runs on this box as root; no VM, VLAN or egress policy |
+| **O**, **X** | none | No container image yet |
-| **O** `nisse-ops` | none | |
+| `specs/` | done for now | `orchestrator.qnt` (current loop), `loop.qnt` (the architecture, #91) |
 | **X** Deployment model | none | |
 ---
 ## 2. The environment today
-**Supporting services** (X5) are installed and running: Matrix (Synapse), Gitea,
+- **Supporting services** (X5): Matrix, Gitea, LiteLLM, Vaultwarden, LLDAP with
-LiteLLM, Vikunja, Outline, Vaultwarden, LLDAP with Authelia. Per-agent accounts and
+  Authelia. The infrastructure repository exists in Gitea.
-tokens are created as each agent needs them. So far `nisse-dev` has a Matrix test
+- **Local inference:** llama-swap on an RTX 4090 host (320 W cap). The benchmark
-identity and room, a Gitea test account and sandbox repo, and dedicated LiteLLM
+  (PR #55) chose `qwen3.8-27b` for local work.
-keys.
+- **Models by role today:** executor `qwen3.8-27b`, planner and helper
-
+  `claude-opus-5-5`, expert `claude-haiku-4-5`. Phase 2 moves to driver and helpers.
-**Local inference:** an RTX 4090 host, `inference-4090.garrett-hall.com`, running
+- **Bring-up** runs on this development VM (`scripts/bringup.sh`), against test
-llama-swap (one model resident at a time), capped at 320 W. The executor benchmark
+  resources: the sandbox repository, the test Matrix room, a budget-capped LiteLLM
-(PR #55) chose `qwen3.8-27b` for `nisse-dev`'s local work.
+  key; configuration in `/root/nisse-e2e/`.
-
+- **A build VM for bring-up** is provided by the operator by hand until Terraform
-**Models by role today** (`nisse-dev.example.yaml`): executor `qwen3.8-27b`, planner
+  (Phase 6); the agent reaches it over SSH with a key from `/root/nisse-e2e/ssh/`.
 and delegate `claude-opus-5-5`, consultant `claude-haiku-4-5`. Phase 2 moves to the
 architecture's roles (L9): driver local, helpers frontier.
 **Bring-up runs** happen on this box, against test resources: the sandbox repo, the
 test Matrix room, a budget-capped LiteLLM key; configuration in `~/nisse-e2e/`.
 ---
 ## 3. How we work
- **One PR per milestone,** usually. Every PR still starts with a short
+- **One PR per milestone,** each starting with a short implementation plan per
-  implementation plan per `CLAUDE.md`. For a **pre-approved** milestone that plan is
+  `CLAUDE.md`. A milestone marked **needs discussion** starts with a design
-  a quick confirmation; a **needs discussion** milestone starts with a design
+  discussion first.
-  discussion before any plan.
+- **Every change comes with tests,** through `scripts/check.sh`.
- **Every change comes with tests,** per `CLAUDE.md`, and runs through
+- **A live run follows each milestone that changes behaviour,** not just each phase:
-  `scripts/check.sh` (lint, format, unit and integration tests).
+  a short bring-up run of what the milestone added. Findings are filed as issues and
- **The Quint model moves with the loop.** Any change to how the loop enters or
+  either fixed before moving on or deliberately deferred.
-  leaves a waiting point updates `specs/`, and every counterexample is reproduced
+- **The Quint models move with the loop:** `orchestrator.qnt` while `control.py`
-  against the real code before it is fixed.
+  runs, `loop.qnt` for the new loop, whose properties become tests as each piece
- **A bring-up run on this box ends each phase** where it can be run live. What it
+  lands (Phase 2).
-  finds is filed as issues and fixed before the next phase starts, or explicitly
+- **`control.py` gets fixes only,** and only for what blocks a live run. New
-  deferred.
+  behaviour goes into the new loop.
- **Bring-up runs on this development VM,** as root, like production. Guards: it
+- **Bring-up runs on this box as root,** against test resources only, with today's
-  touches only test resources, and keeps today's structural gates on in its
+  gates on until the classifier and trust dial are proven (Phase 3).
-  bring-up configuration until the classifier and trust dial are proven (Phase 3).
+- **This plan is revised as we go,** in the same PR as the work that changes it.
 - **This plan is revised as we go:** a milestone's scope may move once work shows
  its real shape, recorded here in the same PR.
 ---
 ## 4. Phases
-### Phase 0 — Housekeeping
+### Phase 0 — Housekeeping (done)
-**M0.1 Retire the old spec.** *Pre-approved.* **Done.**
+**M0.1 Retire the old spec.** **Done.** **M0.2 A bring-up script.** **Done.**
 - Update every section reference in code, tests, `specs/` and the `CLAUDE.md` and
  `README.md` files from the old spec's A–H to the new letters.
 - Delete `docs/Architecture-Build-Spec.md` (git keeps it), and the old-to-new
  section table this plan carried for the purpose.
 - Acceptance: `grep` finds no reference to the old spec or its section numbers.
-**M0.2 A bring-up script.** *Pre-approved.* **Done.**
+### Phase 1 — Execution over SSH, and the agent in a container
 - `scripts/bringup.sh`: set aside the previous run's state, workspace and log;
  export the ref under test; start the agent as a systemd unit, so it outlives the
  shell that started it; collect the log and a readable transcript
  (`python -m nisse_dev.dump`) afterwards.
 - The agent runs as root on this development VM, as it will on its own VM in
  production (D7), so installing packages works as it will there.
 - Acceptance: a smoke run starts, answers `status`, stops cleanly, and `collect`
  produces the log and transcript.
-### Phase 1 — Foundations the new loop needs
+**M1.1 Work items in the state store.** **Done.**
-These fix problems found in the live runs that **survive the loop change.** Issues
+**M1.2 Transcript views.** **Done.** (#78 closes when a live run confirms it.)
 that only exist because of `control.py`'s state machine (#61, #79, #81) are folded
 into Phase 2's design instead of being fixed twice.
-**M1.1 Work items in the state store.** *Needs discussion* (schema). **Done.**
+**M1.3 Execution over SSH.** *Pre-approved.*
- A work-items table with an active pointer (L1); every canonical event carries its
+- `execute(host, command, *, secrets, stdin)` as the only way to run anything; file
-  work item and its actor (L7). A migration for existing stores.
+  tools over SSH; the host bound by the harness from the project (L13).
- Acceptance: two work items back to back keep separate state and events; a
+- The workspace lives on the build VM: cloned there, with no token in its
-  restart keeps both; migration tests.
+  configuration; the push key handed to each push only (D4, #89).
 - Every secret handed out is redacted from tool output (L12).
 - Tests against a throwaway `sshd` container.
 - Acceptance: the existing executor and delegation tests pass over SSH; a live run
  of one milestone on the operator's build VM.
-**M1.2 Transcript views.** *Needs discussion* (rendering). **Done.**
+**M1.4 Projects** (D8). *Pre-approved.*
- Per-role rendering (L6): a role sees only its work item, attributed entries, and
+- The project entry in configuration: repository, default branch, build VM, label
-  orchestrator `fact` events; never another role's raw turns as its own (#78).
+  defaults, secrets by name with their visibility, egress additions.
- Acceptance: the second live run's confusion can't happen (a planning call after
+- Acceptance: the bring-up configuration is one project entry; a second project
-  a milestone contains the milestone fact and PR, and no executor turns).
+  needs only a second entry.
-**M1.3 A fresh workspace at `use`, and harness-owned branches.** *Pre-approved.*
+**M1.5 The agent as a container image.** *Pre-approved.*
- Fetch; clean checkout of the latest base; set aside anything left over and report
+- A Dockerfile for `nisse-dev`, stateless with a volume for state; a Gitea Actions
-  it (D3). Milestone branches cut by the harness from the task's workspace;
+  workflow that builds it and pushes to Gitea's registry on each release tag (X4).
-  "changes to publish" counted from the milestone's start (D4). Covers #62 and the
+- `bringup.sh` runs the image instead of a venv.
-  clone-on-use half of #27.
+- Acceptance: a tagged release produces an image; a live run from the image.
 - Acceptance: a leftover branch, uncommitted change and stale `origin` are each
  handled and reported; a second milestone's PR contains only its own commits.
-**M1.4 Publish only the work.** *Pre-approved.*
+**Phase 1 acceptance:** a live run, from the image, of a milestone and a review
- Standard excludes in the clone, and commit only what the work changed (D4, #80).
+comment on the build VM.
 - Acceptance: a milestone that ran pytest publishes no `.pyc`; excluded files are
  reported.
-**M1.5 Review intake.** *Pre-approved.*
+### Phase 2 — The new loop, as a thin slice first
 - Inline review comments and review bodies (#69); no stall when a watch is
  replaced from its own callback (#70); replies in the thread where the comment was
  made (D5).
 - Acceptance: tests for each; inline comments work in the phase's bring-up run.
-**Phase 1 acceptance:** a bring-up run on the current loop with two milestones and
+Built alongside `control.py`, reusing the state store, log, views, publisher,
-a review comment, including inline ones.
+pollers and recovery. The first milestone is the smallest end-to-end path; each
 later one widens it and is followed by a live run.
-### Phase 2 — The conversational loop
+**M2.1 The slice.** *Needs discussion* (the loop's shape in code).
 - An issue assigned to `dev-agent` → a work item (D1); a restatement in the item's
  thread; a go-ahead bound to its revision (L3, L5); one milestone through
  `start_milestone` / `finish_milestone` on a harness branch (D4); the PR; a review
  comment answered (D5); the merge; the item closed.
 - Includes what carries over: a fresh workspace at item start and harness-owned
  branches (D3, D4, #62); publishing only the work (#80); review intake, inline
  comments and the watch stall (#69, #70); the review-fix check after merge.
 - `loop.qnt`'s properties for these paths become tests.
 - Acceptance: a live run of that path, plus a crash mid-publish.
-Built **alongside** `control.py`, reusing the state store, canonical log, context
+**M2.2 The queue.** *Pre-approved.* Several items: states, park-and-switch with
-fitting, publisher, pollers and recovery; `nisse-dev` switches over when the new
+branches pushed, next-item order, blocked items from Gitea dependencies, a closed
-loop passes its tests and a bring-up run, and `control.py` is retired.
+PR handled (L1, D4). Live run with two issues and a dependency.
-**M2.1 Loop design and its model.** *Needs discussion.*
+**M2.3 Threads and labels.** *Pre-approved.* Routing by thread, ambiguous commands
- A short design note (in the PR, then folded into the architecture where it
+asked about (L5); `bug` grace periods, `discuss`, `design`; questions and
-  settles something): the loop's turns, waiting points, pending questions,
+plain-language go-aheads (D1). Live run.
  permissions it checks, and recovery. A new Quint model for it before code, with
  the no-stranding, approval-binding and at-least-once properties carried over.
-**M2.2 The loop core.** *Needs discussion.*
+**M2.4 Multi-milestone work and documents.** *Pre-approved.* Build plans in the
- The driver-driven turn loop (L2): conversation, tool calls, pending questions,
+repository, `ARCHITECTURE.md` kept current, waiting for merges between milestones
-  `stop`, `abandon`, intents and boot recovery on the work-items schema.
+(D2, D4). Live run of a two-milestone feature.
 - **Questions never change state** (L5): the recognition rule is decided here (Q).
 - Acceptance: the new Quint model's scenarios reproduced as tests; no-stranding
  under injected faults, as `test_no_stranding.py` does today.
-**M2.3 Roles and capability profiles.** *Pre-approved.*
+**M2.5 Conversations and new projects.** *Needs discussion.* Work started in
- Roles become `driver`, `helper`, `expert`, `classifier` (L9); capability profiles
+conversation, record issues carrying their item's ID, the new-project flow and its
-  in configuration, filled from the benchmark; the driver's prompt encourages asking
+project entry (D1, D8, #29). Live run of a small new project.
  for help; observable escalation triggers; delegation and `ask_expert` wired to the
  new loop. Supersedes #46 (separate architect and planner roles).
-**M2.4 Conversation.** *Pre-approved* (the behaviour is set by L5).
+**M2.6 Roles and capability profiles.** *Pre-approved.* `driver`, `helper`,
- Explanations with plans and revisions, finished-work summaries, progress in prose
+`expert`, `classifier` (L9); escalation triggers; delegation and `ask_expert` on the
-  (#42), handling messages that can't be acted on yet, asking mid-work (#77).
+new loop.
-**M2.5 Documents as the workflow.** *Needs discussion.*
+**M2.7 Conversation.** *Pre-approved.* Explanations, progress in prose, asking
- Project `ARCHITECTURE.md` and build plans in the repo, discussed and landed by PR;
+mid-work (L5, #42, #77).
  milestones marked pre-approved or needs discussion and whether they change code;
  permissions derived from them (L3, D2).
 - Milestones and projects end properly: the plan declares its milestones; a
  no-change milestone that wasn't meant to change code is done; moving on checks the
  PR's state; finishing closes the issue (D4, D5). Supersedes #61, #79, #81.
-**M2.6 Switch over and retire `control.py`.** *Pre-approved, after acceptance.*
+**M2.8 Switch over.** *Pre-approved, after acceptance.* `nisse-dev` runs the new
- `nisse-dev` runs the new loop; `control.py`, its tests and the old Quint model are
+loop; `control.py`, its tests and `orchestrator.qnt` are removed.
  removed.
 **Phase 2 acceptance:** a bring-up run: a new small project from conversation to an
 agreed `ARCHITECTURE.md` and build plan, two milestones (one pre-approved), a review
 comment answered, a question asked by the agent mid-milestone, and a crash
 mid-review.
 ### Phase 3 — Authority
-**M3.1 Action classes and the fast path.** *Needs discussion.*
+**M3.1 Action classes and the fast path.** *Needs discussion.* Harmless, contained
- Every tool action classified by effect (P3); static fast-path rules per agent
+and dangerous actions per agent (P3, D6), static fast-path rules.
  (D6); the scope each class-2 action is checked against.
-**M3.2 The action classifier.** *Needs discussion.*
+**M3.2 The classifier.** *Needs discussion.* L4, its model, fail-closed verdicts,
- L4, including the model choice (Q), fail-closed behaviour, logged verdicts, and
+overrides as an evaluation set.
  operator overrides as an evaluation set.
-**M3.3 Permissions and the trust dial.** *Needs discussion.*
+**M3.3 The trust dial.** *Needs discussion.* Per project (D8); agent merging behind
- Approvals as scoped permissions, the operator-set trust dial, and agent merging
+protection it can't satisfy, verified by a test (D4, D5, #65).
  behind branch protection the agent can't satisfy, verified by a test (L3, D4, D5,
  #65).
 **Phase 3 acceptance:** a bring-up run with the classifier active and gates first
 on, then deliberately loosened in the bring-up configuration.
 ### Phase 4 — The environment
-**M4.1 Toolchain preflight.** *Pre-approved.* Detect what the project needs, check
+**M4.1 Toolchain preflight, M4.2 recording installs, M4.3 environment repair.**
-it, and install what's missing before planning; toolchain facts in the transcript
+*Pre-approved.* All on the build VM (D3).
 (D3).
-**M4.2 Recording installs.** *Pre-approved.* Every package or toolchain the agent
+### Phase 5 — Memory
 installs is recorded as a fact and in the project's scope memory, so the
 environment can be rebuilt; recurring ones are noted for the VM image (D3, X4).
-**M4.3 Environment repair.** *Pre-approved.* The driver may clean caches and
+**M5.1 The vault.** *Pre-approved.* The folder, per-agent mounts, control tools,
-reinstall project dependencies within a task, and reports it (D3).
+atomic writes and provenance (L10).
-### Phase 5 — Memory and knowledge
+**M5.2 Agent memory** (*needs discussion*) and **M5.3 scope memory**
 (*pre-approved*): when notes are written and read, verbatim corrections (L6).
-**M5.1 The Outline client.** *Pre-approved.* Per-agent Outline user and token,
+**M5.4 The operator's copy.** *Pre-approved.* Syncthing, directions per folder.
 collections and documents by ID, revision reads, one writer per document (L10).
-**M5.2 Agent memory.** *Needs discussion.* When notes are written and read, the
+### Phase 6 — Production
 index, verbatim corrections (L6).
-**M5.3 Scope memory for repositories.** *Pre-approved.* Project documents read at
+**M6.1 The agent host.** *Needs discussion.* Containers, volumes, snapshots,
-work-item start and snapshotted; summaries at close (L6).
+per-agent networks (X1, X2, X4).
-### Phase 6 — `nisse-dev` in production
+**M6.2 Build VMs by Terraform.** *Needs discussion.* The template, VLAN and egress
 (D7, X2), collection when idle.
-**M6.1 The VM, VLAN and egress policy.** *Needs discussion.* D7, X1, X2.
+**M6.3 Secrets and identity.** *Pre-approved.* Per-agent secret sets on the host,
 LLDAP principal (L12).
-**M6.2 Secrets and identity.** *Pre-approved.* Vaultwarden machine accounts, LLDAP
+**M6.4 Delivery.** *Pre-approved.* Watchtower, restarts gated on an idle flag (X4).
 principal, secrets out of `.env` files (L12).
-**M6.3 Images and delivery.** *Pre-approved.* Gitea Actions → OCI registry →
+**M6.5 The browser tool.** *Pre-approved.* On the internal test network (D7).
 Watchtower, restarts gated on an idle flag (X4).
 **M6.4 The browser tool.** *Pre-approved.* Playwright on the internal test network
 only (D7).
 ### Phase 7 — `nisse-core` and `nisse-ops`
-**M7.1 Extract `nisse-core`.** *Pre-approved.* A package split within the monorepo:
+**M7.1 Extract `nisse-core`.** **M7.2 MCP servers** (#37). **M7.3 Approved step
-what both agents use moves into `nisse-core`.
+sequences** (O2). **M7.4 Ops intake and duties** from the infrastructure
-
+repository (O5). **M7.5 The ops tool VM and deployment** (O6). **M7.6 Requests
-**M7.2 MCP servers.** *Needs discussion.* Per (service, site, privilege) instances,
+between agents:** `nisse-dev` asking for build VMs (L1, O5). All *needs discussion*
-tool profiles, configuration (X3, #37).
+except M7.1.
 **M7.3 Approved step sequences.** *Needs discussion.* The ops plan format, matcher,
 policy file, expiry (O2).
 **M7.4 Tickets and duties.** *Needs discussion.* Vikunja intake and approvals,
 scheduled checks and alert intake as duties (O5, #33).
 **M7.5 `nisse-ops` deployment.** *Needs discussion.* O6.
 ### Not yet scheduled
-From the architecture's open questions (Q) and older issues, to be placed when
+From the architecture's open questions and older issues: filing issues from
-their trigger arrives:
+conversation; stacked PRs; releases; `nisse-ops` issuing credentials; a server
- more than one active work item per agent;
+database; recovering the agent host (an ops issue); more than one active work item;
- vision for 3D design and video work;
+vision; one local model for every agent; several forges (#26); a room per project
- one local model for every agent, or agents taking turns on the inference host;
+(#28).
 - several forges and projects (#26); a Matrix room per project (#28); creating a
  project from scratch and who owns it (#29); one agent instance per project (#30);
 - infrastructure as code for agent VMs.
 ---
 ## 5. Open issues and where they land
-All 19 open issues, as of 2026-09-30:
+| Issue | Lands in |
- **11 land in a milestone** and stay open until it is done.
+|---|---|
- **4 are superseded** by the new loop's design. They only exist because of
+| #26 multiple forges | Not yet scheduled |
-  `control.py`'s state machine, so fixing them there would be thrown away. They are
+| #27 per-project state, clone on use | M1.3 (clone on the build VM), M1.4 (projects) |
-  closed, with a pointer to their milestone, when this plan is agreed.
+| #28 room per project | Not yet scheduled (threads, M2.3, may make it unnecessary) |
- **4 are not yet scheduled** and stay open until their trigger arrives.
+| #29 create a project from scratch | M2.5 (ownership settled by D4, D8) |
-
+| #37 MCP servers | M7.2 |
-| Issue | Status | Lands in |
+| #42 per-turn commentary | M2.7 |
-|---|---|---|
+| #62 branch at execution start | M2.1 |
-| #26 multiple forges | Not yet scheduled | |
+| #65 agent merges approved PRs | M3.3 |
-| #27 per-project state, clone on use | Milestone | M1.1 (state), M1.3 (clone) |
+| #69 inline review comments | M2.1 |
-| #28 room per project | Not yet scheduled | |
+| #70 watch stall | M2.1 |
-| #29 create a project from scratch | Not yet scheduled | Ownership decision first |
+| #77 conversational agent | M2.3 (questions), M2.7 |
-| #30 agent instance per project | Not yet scheduled | |
+| #78 transcript scoping and attribution | Done in M1.1, M1.2; closes after a live run |
-| #33 Vikunja tickets | Milestone | M7.4 |
+| #80 build artifacts in PRs | M2.1 |
-| #37 MCP servers | Milestone | M7.2 |
+| #89 token in the clone's `.git/config` | M1.3 |
 | #42 per-turn commentary | Milestone | M2.4 |
 | #46 architect and planner roles | **Superseded** | M2.3 (one driver, helpers) |
 | #61 simpler milestone gating | **Superseded** | M2.5 |
 | #62 branch at execution start | Milestone | M1.3 |
 | #65 agent merges approved PRs | Milestone | M3.3 |
 | #69 inline review comments | Milestone | M1.5 |
 | #70 watch stall | Milestone | M1.5 |
 | #77 conversational agent | Milestone | M2.2 (questions), M2.4 |
 | #78 transcript scoping and attribution | Milestone | M1.1, M1.2 |
 | #79 ending projects and milestones | **Superseded** | M2.5 |
 | #80 build artifacts in PRs | Milestone | M1.4 |
 | #81 no-change milestones | **Superseded** | M2.5 |
 ## 6. Implementation notes