ultra-builder-pro-cli 0.26.0 → 0.26.2

package/.ultra-template/contexts/TEMPLATE.md

@@ -32,6 +32,8 @@
 <!-- The verifiable definition of done. Read by mid_workflow_recall and session_context, and
      injected into agent context — keep every line testable and keep noise out. -->
 
+**Change Acceptance IDs**: [`AC-1`, `AC-3`, or `none` for task-local-only work]
+
 - [ ] `exact test command`
 - [ ] Pass: [scenario 1]
 - [ ] Pass: [scenario 2]
@@ -51,6 +53,11 @@
 
 **Source**: `.ultra/specs/product.md#section-id`
 
+**Hard Constraints**: [`HC-1`, `HC-3`, or `none` — the north-star constraints this
+task could plausibly violate. Read them before the first edit and again before
+marking complete. Crossing one stops the work and goes to the owner, however good
+the argument.]
+
 **Decisions**: [relevant `.ultra/decisions/<id>.md` entries, or `none`]
 
 ## Change Log

package/.ultra-template/north-star.md

@@ -1,46 +1,35 @@
 # Project North Star
 
-> **Authority**: the owner's literal intent for this project, kept short enough to re-inject every
-> session. Every task must trace back here.
+> **Authority**: the accepted steering contract established by `ultra-research` from
+> owner decisions and cited evidence. Every Change and task must serve this contract.
 >
-> **Scope**: only what cannot be derived from `specs/product.md`. Success metrics,
-> excluded features, actors and scenarios stay in that file — do not copy them here.
+> **Scope**: this file owns the project direction, the one North Star outcome or metric
+> when one is justified, hard constraints, and explicit exclusions. Product requirements,
+> supporting measurements, actors, and scenarios stay in `specs/product.md`.
 
----
+## Project Direction
 
-## One-line
+[NEEDS CLARIFICATION]
 
-<!-- ONE sentence, in the owner's own words, describing what they want.
-     Captured by ultra-init, or from the owner's first request.
-     Do not paraphrase into product-speak: specs/product.md holds the processed version.
-     The value of this line is that it is the unprocessed baseline you check the
-     processed version against. -->
+## North Star Outcome
 
-_(not yet defined — run `ultra-init`, or the first owner request will populate this)_
+- `NS-01` outcome: [NEEDS CLARIFICATION]
+- Metric or observation contract: [NEEDS CLARIFICATION]
+- Why it represents delivered value: [NEEDS CLARIFICATION]
 
----
+A single metric is optional. When one would distort the product, record that decision and
+use an observable outcome plus guardrails instead of inventing numeric precision.
 
 ## Hard Constraints
 
-<!-- What must never happen, even when convenient. These are prohibitions, not technical
-     constraints — architecture.md holds those. Examples:
-     - Never store plaintext passwords
-     - Cannot break backwards compatibility with API v1
-     - Bundle stays under 500 KB
-     - External API cost stays under $X/month -->
+- `HC-1`: [NEEDS CLARIFICATION]
 
-_(not yet defined)_
+## Explicit Exclusions
 
----
+- [NEEDS CLARIFICATION]
 
-## Notes for agents
+## Research Trace
 
-1. Re-anchor on the One-line — that is the literal owner request. When the spec and this line
-   disagree, the spec drifted.
-2. Check Hard Constraints before any "improvement". An improvement that violates a constraint is
-   a regression.
-3. If your current work does not trace back here, **stop and ask**. You may be building the
-   wrong thing.
-
-The One-line is injected by `session_context` at SessionStart. `mid_workflow_recall`
-re-injects the active task acceptance before relevant tool use. See PHILOSOPHY C1.
+- Project brief: `project-brief.md`
+- Accepted research run: [NEEDS CLARIFICATION]
+- Decision references: [NEEDS CLARIFICATION]

package/.ultra-template/project-brief.md

@@ -0,0 +1,28 @@
+# Project Brief
+
+> **Authority**: raw owner intake captured by `ultra-init`. It preserves what the owner
+> initially asked for and the broad outline needed to begin research. It is not an
+> evidence-backed product baseline.
+>
+> **Lifecycle**: `ultra-research` reads this seed and promotes accepted conclusions into
+> the North Star, specifications, domain language, or decisions. It does not rewrite the
+> brief to make later conclusions look inevitable.
+
+## One-line
+
+[NEEDS CLARIFICATION]
+
+## Initial Outline
+
+- What it is: [NEEDS CLARIFICATION]
+- Broadly for whom or what: [NEEDS CLARIFICATION]
+- Intended result: [NEEDS CLARIFICATION]
+- Broad boundary: [NEEDS CLARIFICATION]
+
+## Explicit Inputs
+
+- [NEEDS CLARIFICATION]
+
+## Open Questions for Research
+
+- [NEEDS CLARIFICATION]

package/.ultra-template/specs/product.md

@@ -60,6 +60,10 @@ Scope status must reflect an explicit decision. Do not infer a reduced product p
 
 ## 22 Outcomes and Measurements
 
+The accepted project direction and any single North Star outcome or metric live in
+`.ultra/north-star.md`. Record supporting outcomes, measures, and guardrails here; do
+not force one metric when an observable outcome plus guardrails is the honest contract.
+
 | Outcome or guardrail | Definition | Evidence source | Decision use | Owner |
 |---|---|---|---|---|
 | | | | | |

package/.ultra-template/specs/research-distillate.md

@@ -28,6 +28,14 @@ specifications' prose.
 |---|---|---|
 | | | |
 
+## Source Revisions
+
+| Canonical source | Git blob hash from `git hash-object` | Status |
+|---|---|---|
+| `.ultra/specs/product.md` | | current or stale |
+| `.ultra/specs/architecture.md` | | current or stale |
+| `.ultra/specs/discovery.md` | | current or stale |
+
 ## End-to-End Trace
 
 | Problem or constraint | Actor and scenario | Requirement and acceptance | Architecture path | Verification |

package/.ultra-template/test-report.json

@@ -3,6 +3,7 @@
   "change_id": null,
   "task_ids": [],
   "git_commit": null,
+  "intent_digest": null,
   "worktree": {
     "dirty": null,
     "diff_digest": null

package/CHANGELOG.md

@@ -7,6 +7,125 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.26.2] — 2026-08-04
+
+### Added
+
+- Added `.ultra/project-brief.md` as the raw owner-intake artifact and optional
+  Research `wayfinding.md` navigation for unclear multi-lens paths, without adding a
+  fifteenth Skill or another semantic authority.
+- Expanded Research alternatives beyond direct competitors and made a single North
+  Star metric optional when an observable outcome plus guardrails is more honest.
+
+### Changed
+
+- Narrowed `ultra-init` to the Project Brief and empty skeleton, made
+  `ultra-research` own the first accepted North Star, domain language, and
+  specification baseline, and limited `ultra-change` to touched baseline sections.
+- Made the first six Research lenses dependency-aware and kept Grilling, Think, and
+  Domain Modeling as caller-bounded reusable methods rather than Research stages or
+  owner routes.
+- Routed `ultra-status` from semantic file contents instead of a Research completion
+  bit or any single unresolved marker.
+- Gave every Change one stable id across active, archive, and abandoned positions;
+  made the task ledger append-only and every current reader active-id scoped.
+- Added explicit `none`, `bounded`, and `required` Research Disposition contracts,
+  one-task planning for quick Changes, and two-pass delivery against a fresh semantic
+  plus product snapshot.
+- Gave the exact Change Contract explicit Profile rationale and Risk flags fields so
+  its required blast-radius decision has one stable representation.
+- Kept ordinary technical seam selection model-owned and made delegation reachable for
+  task execution, scoped Research evidence, and aggregate Change review without
+  manufacturing task rows.
+
+### Fixed
+
+- Made session context prefer the accepted North Star, fall back to the Project Brief
+  before Research, retain legacy one-line compatibility, and always append active task
+  acceptance when present.
+- Bound Research distillates to the actual `git hash-object` value of all three
+  baseline specifications so staleness is mechanically observable.
+- Prevented archived or abandoned unfinished tasks, invalid Change ids, blocked
+  dependencies, and ambiguous multiple `in_progress` rows from being injected as the
+  current Hook acceptance.
+- Made canonical `change_id` override a conflicting legacy `change_ref`, so migration
+  compatibility cannot revive a historical task.
+- Bound Test reports to exact Change id, ordered task ids, intent digest, HEAD, and a
+  product-worktree digest that excludes the report and Change-directory metadata while
+  binding current intent separately, so product reconciliation requires retest while
+  delivery metadata and archive finalization do not self-invalidate.
+- Prevented a separate request from opening a second active Change, and gave abandoned
+  intents an exact owner decision, reason, reusable-evidence, and recovery closure read
+  by future Change and Status workflows.
+- Kept non-publishing package inspection diagnostic while the explicit Deliver gate
+  blocks release-package creation only for an undispositioned changed export with no
+  non-test consumer.
+- Made that gate's authorization path use the finding disposition in the current Test
+  report and carry the accepted decision into `delivery.md`.
+- Aligned `.npmignore` with the explicit package allowlist and locked all eight
+  maintained product documents into the tarball smoke contract.
+
+### Verified
+
+- Passed 119 Node tests, 21 Hook tests, 14 Skill Creator validations, five isolated
+  host install/Doctor/uninstall lifecycles, a 115-entry package dry-run inspection, and
+  an audit with zero vulnerabilities.
+
+## [0.26.1] — 2026-08-03
+
+### Added
+
+- Added reframing to `ultra-grilling`, five bounded framing questions, and checkable
+  resolved, stalled, and unavailable exits so a literal request can be translated
+  into the role and outcome the owner actually means.
+- Added `HC-<n>` hard-constraint identifiers, task-level constraint mapping,
+  executable Change verification, and a supported `changes/abandoned/` exit.
+- Added owner-authorized cross-model-family review for major or security-relevant
+  Changes, plus optional native fan-out inside independent research and test regions.
+
+### Changed
+
+- Made `ultra-dev` converge on each task's mapped Change acceptance IDs, track the
+  best-ever passing set, and stop after two no-progress rounds or three repair rounds.
+- Expanded delivery reconciliation to every repository document and kept its wiring
+  gate local to an explicitly invoked `ultra-deliver`; ordinary npm lifecycle commands
+  never depend on project `.ultra/` evidence.
+- Made `ultra-change` read relevant archive history before opening intent, preferring
+  `delivery.md` while supporting legacy archive summaries and verification records.
+- Made all six review lenses the default and required every skipped lens to preserve
+  its reason in the aggregate result.
+
+### Fixed
+
+- Prevented delegated workers from modifying `.ultra/`, including when `.` grants the
+  rest of an isolated checkout.
+- Made dangerous-command authorization reachable by hashing the protected effect
+  independently of its retry spelling.
+- Kept additive protected-branch publication on the host's native authority path instead
+  of hard-blocking an already owner-authorized release, retained exact-digest guards for
+  history rewrites and branch deletion, and stopped treating inert searches as effects.
+- Kept expandable interpreter heredocs inside dangerous-effect inspection while
+  continuing to treat quoted commit-message heredocs as data.
+- Required explicit owner authority before `ultra-review` can launch a cross-host
+  model-family recheck.
+- Bound `ultra-dev` convergence to the active Change acceptance IDs, with explicit
+  stalled and unreachable exits instead of an open-ended repair loop.
+- Made `ultra-change` recover archived intent from legacy archive evidence when the
+  preferred delivery record is absent.
+- Kept the `ultra-deliver` wiring gate local to an owner-invoked delivery workflow;
+  ordinary package commands remain independent of project `.ultra/` state.
+- Added the missing `changes/abandoned/` template directory and migrated completed WIP
+  evidence links to their canonical runtime document.
+
+### Verified
+
+- Completed an authenticated Claude-to-Codex continuation on one isolated
+  file-first task: Claude banked the failing test and Resume Note without changing
+  implementation, and Codex resumed from Git plus `.ultra/`, reached green, and
+  closed the ledger and canonical evidence without conversation transfer.
+- Verified the release candidate with 111 Node tests, 10 Hook tests, 14 Skill Creator
+  validations, five healthy host Doctor reports, and zero audit vulnerabilities.
+
 ## [0.26.0] — 2026-08-01
 
 ### Added
@@ -1167,7 +1286,18 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   skill manifest, CLI protocol + mapping table; 5 spec validators.
 - **Phase 0 — skeleton**: multi-runtime installer scaffolding.
 
-[Unreleased]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.20.0...HEAD
+[Unreleased]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.26.1...HEAD
+[0.26.1]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.26.0...v0.26.1
+[0.26.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.25.1...v0.26.0
+[0.25.1]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.25.0...v0.25.1
+[0.25.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.24.2...v0.25.0
+[0.24.2]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.24.1...v0.24.2
+[0.24.1]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.24.0...v0.24.1
+[0.24.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.23.0...v0.24.0
+[0.23.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.22.0...v0.23.0
+[0.22.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.21.1...v0.22.0
+[0.21.1]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.21.0...v0.21.1
+[0.21.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.20.0...v0.21.0
 [0.20.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.19.1...v0.20.0
 [0.19.1]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.19.0...v0.19.1
 [0.19.0]: https://github.com/rocky2431/ultra-builder-pro-cli/compare/v0.18.0...v0.19.0

package/README.md

@@ -29,14 +29,14 @@ The package installs exactly fourteen Skills in three roles.
 
 | Skill | Outcome |
 |---|---|
-| `ultra-init` | Establish the project skeleton and one-line north star |
-| `ultra-research` | Verify external claims and refresh specifications |
-| `ultra-change` | Reconcile a requested change against current product truth |
+| `ultra-init` | Establish the project skeleton and raw Project Brief |
+| `ultra-research` | Establish an accepted North Star and evidence-backed baseline |
+| `ultra-change` | Reconcile one requested delta against the accepted baseline |
 | `ultra-plan` | Write tracer-bullet tasks, contexts, dependencies, and seams |
 | `ultra-dev` | Implement one task through red/green development and six evidence dimensions |
 | `ultra-test` | Audit whole-system wiring, E2E behavior, performance, and security |
 | `ultra-deliver` | Reconcile, review, document, and archive a completed Change |
-| `ultra-delegate` | Run a bounded instruction through another supported CLI |
+| `ultra-delegate` | Run bounded task work, Research evidence, or aggregate review through another supported CLI |
 
 ### Model-invoked disciplines
 
@@ -60,6 +60,14 @@ review, and a consequential unresolved decision uses think. A host with native b
 subagents may parallelize review lenses; another host runs the same lens assets
 sequentially.
 
+Inside `ultra-research`, the first six references are six semantic lenses, not six
+extra Skills: problem validation, opportunity discovery, market assessment,
+alternatives, product strategy, and assumptions validation. `wayfinding.md` chooses
+the smallest dependency-correct path through those and the later evidence lenses.
+Grilling still owns how to ask one missing question, Think owns one consequential
+trade-off, and Domain Modeling owns vocabulary; Research owns the overall question
+map, evidence convergence, owner checkpoints, and baseline promotion.
+
 ### What happened to the original Agents
 
 v0.26 does not install a custom `agents/` projection. The old review workers became the
@@ -72,12 +80,15 @@ host exposes the same custom-agent API.
 
 ## Project authority
 
-A new project uses this shape:
+A project after its first accepted Research baseline uses this shape. Immediately after
+Init, `CONTEXT.md` does not exist yet and the North Star/specification files are empty
+skeletons.
 
 ```text
 CONTEXT.md
 .ultra/
 ├── .gitignore                  # ignores only derived Ultra paths
+├── project-brief.md
 ├── north-star.md
 ├── tasks.json
 ├── test-report.json
@@ -88,15 +99,18 @@ CONTEXT.md
 │   └── research-distillate.md
 ├── changes/
 │   ├── active/<change-id>/{intent.md,delivery.md}
-│   └── archive/<change-id>/intent.md
+│   ├── archive/<change-id>/{intent.md,delivery.md}
+│   └── abandoned/<change-id>/intent.md  # includes exact Abandonment closure
 ├── contexts/<task-id>.md
 ├── decisions/<decision-id>.md
 ├── evidence/<task-id>/...
-└── research/<run-id>/...
+└── research/<run-id>/{brief.md,<step-id>.md}
 ```
 
-Canonical semantic facts live in those files. Git provides history, comparison,
-rollback, and archive moves. The following are derived and can be deleted or rebuilt:
+The optional `brief.md` is derived navigation; the selected step reports are cited
+evidence, and promoted semantic facts live in the other canonical files. Git provides
+history, comparison, rollback, and archive moves. The following additional paths are
+derived and can be deleted or rebuilt:
 
 ```text
 .ultra/.runtime/
@@ -111,13 +125,13 @@ staleness, and recovery rules.
 
 | Route | Canonical write | Who consumes it next |
 |---|---|---|
-| `ultra-init` | north star and four specification skeletons; empty task and test ledgers | every later workflow and context Hook |
-| `ultra-research` | selected cited reports plus reconciled specifications and distillate | change, plan, and delivery |
-| `ultra-change` | one active `intent.md` with exact acceptance, seams, reconciliation, and recovery | plan, dev, test, review, status, delivery |
-| `ultra-plan` | `tasks.json`, one context per task, and Planning Posture in the active intent | dev and every resume path |
+| `ultra-init` | raw `project-brief.md`, empty North Star and specification skeletons, and empty task/test ledgers | research, status, and the pre-baseline session Hook fallback |
+| `ultra-research` | accepted North Star, first domain baseline, selected cited reports, reconciled specifications, and distillate | change, plan, and delivery |
+| `ultra-change` | one active `intent.md` with stable `change_id`, Research Disposition, and only the accepted baseline sections touched by that delta; or an exact Abandonment closure before an owner-authorized move | active: research through delivery; abandoned: future Change history and status |
+| `ultra-plan` | append-only `tasks.json` rows for the active `change_id`, one context per task, and Planning Posture in the active intent | dev and every resume path |
 | `ultra-dev` | source/tests, task evidence, synchronized task/context status, Completion and Resume Note | review, test, status, delivery |
-| `ultra-test` | the one current `test-report.json` bound to HEAD and worktree digest | status and delivery |
-| `ultra-deliver` | one `delivery.md`, reconciled specs/docs, then the archived Change directory | owner and future Change history |
+| `ultra-test` | the one current `test-report.json` bound to Change id, current task ids, intent digest, HEAD, and product-worktree digest | status and delivery |
+| `ultra-deliver` | first reconciled specs/docs, then after a fresh Test snapshot one `delivery.md` and the archived Change directory | owner and future Change history |
 | `ultra-status` | none | recommends the smallest explicit route from current files |
 | `ultra-delegate` | derived runtime receipt plus an isolated worktree diff | primary host inspection and optional integration |
 
@@ -178,23 +192,37 @@ installation; all five support managed global installation and isolated
 
 ## Typical workflow
 
-1. Select `ultra-init` once. It writes the skeleton, records the owner's one-line goal,
-   initializes Git when needed, and recommends the next route.
-2. Select `ultra-research` when a product or architecture claim needs evidence.
-3. Select `ultra-change` for the requested outcome and reconcile it against current
-   specifications before planning.
-4. Select `ultra-plan` to confirm public seams and produce resumable tracer-bullet tasks.
+1. Select `ultra-init` once. It writes the skeleton, preserves the owner's raw one-line
+   request and broad outline in the Project Brief, initializes Git when needed, and
+   stops before product research.
+2. Select `ultra-research` to turn that brief into the first accepted North Star,
+   shared vocabulary, and evidence-backed product and architecture baseline. For an
+   unclear multi-lens question it first writes a derived Wayfinding brief; for one
+   bounded evidence gap it skips that extra file.
+3. Select `ultra-change` for a requested delta. It reconciles only the baseline sections
+   the delta touches before planning; it does not rebuild the project baseline. Its
+   Research Disposition either cites sufficient evidence or names the bounded question
+   and exit evidence that must return through Research. A micro edit that makes no
+   specification sentence false stays outside Ultra; an accepted quick Change still
+   receives one Plan task. While that Change remains active, reconcile its same stable
+   id; never open a second active Change for a separate request.
+4. Select `ultra-plan` to record public seams and produce resumable tracer-bullet tasks.
+   The model owns ordinary technical seams; the owner decides only a seam that changes
+   the public contract or another material trade-off.
 5. Select `ultra-dev` for one task at a time. Each task leaves a Completion entry,
    evidence, and a closing Resume Note.
-6. Select `ultra-test` once the ledger is complete. Local green tests do not substitute
-   for wiring and E2E proof.
-7. Select `ultra-deliver` to run aggregate review, reconcile documentation, and archive
-   the Change. Commit, push, tag, publication, and deployment remain separately
-   authorized effects.
+6. Select `ultra-test` once tasks for the active Change are complete. Historical ledger
+   rows do not enter this audit. Local green tests do not substitute for wiring and E2E proof.
+7. Select `ultra-deliver` to run aggregate review and reconcile documentation. If that
+   changes product or semantic files, rerun Test; re-enter Deliver on the fresh snapshot
+   to write delivery metadata and archive the stable Change id. Commit, push, tag,
+   publication, and deployment remain separately authorized effects.
 
 At any point, `ultra-status` can reconstruct the current position. A fresh session or
 different host resumes by reading `.ultra/tasks.json`, the selected `context_file`, its
-`## Resume Note`, `CONTEXT.md`, relevant decisions, the active Change, and Git.
+`## Resume Note`, `CONTEXT.md`, relevant decisions, the active Change, and Git. It first
+filters the append-only ledger to tasks whose `change_id` matches the unique active
+Change; archived and abandoned unfinished rows are history, not the frontier.
 
 ## Three independent integration defences
 
@@ -215,11 +243,11 @@ Five optional hooks accelerate file reading and protect a narrow effect boundary
 
 | Hook | Behavior |
 |---|---|
-| `session_context.py` | Inject north star and current acceptance at session start |
+| `session_context.py` | Inject the accepted North Star, or the Project Brief fallback before Research, plus current acceptance |
 | `mid_workflow_recall.py` | Restate acceptance before relevant source operations |
 | `compact_context.py` | Save and restore a disposable Git/file snapshot |
 | `post_edit_guard.py` | Record mechanical evidence observations after edits |
-| `block_dangerous_commands.py` | Deny named destructive shell effects until the exact command is authorized |
+| `block_dangerous_commands.py` | Advise on additive protected-branch publication; deny history rewrites and named destructive shell effects until the exact command is authorized |
 
 All five exit silently when `.ultra/` is absent. Details are in
 [Plugin Isolation](docs/PLUGIN-ISOLATION-CONTRACT.md).
@@ -244,6 +272,10 @@ Use `skills/ultra-delegate/scripts/delegate_wait.py` to wait for `result.json` w
 loading intermediate worker output into the parent context. Delegation grants no new
 authority; writable roots remain inside the worktree, and its permission file must
 declare zero external effects without widening the host sandbox.
+The immutable instruction names task execution/continuation, scoped Research evidence,
+or aggregate Change review/verification. The latter two may be read-only and do not
+require a task row, keeping pre-Plan Research and post-task cross-family review
+reachable without inventing ledger work.
 
 ## Host support
 

package/bin/delegate-worker.cjs

@@ -7,6 +7,7 @@ const path = require('node:path');
 const { execFileSync, spawn } = require('node:child_process');
 
 const RESULT_SCHEMA = 'ultra-delegation-result-v1';
+const PROTECTED_ROOT = '.ultra';
 const MODEL_FIELDS = new Set([
   '$schema', 'status', 'summary', 'changed_files', 'checks', 'evidence',
   'questions', 'residual_risks',
@@ -148,7 +149,11 @@ function validateModelResult(payload) {
   return null;
 }
 
+// The project's memory has exactly one writer: the primary host. A worker that edits
+// `.ultra` in its own worktree collides in the files every later session reads to
+// resume, so no permission grants it — not even `writable_roots: ["."]`.
 function authorized(spec, file) {
+  if (file === PROTECTED_ROOT || file.startsWith(`${PROTECTED_ROOT}/`)) return false;
   return spec.writable_roots.some((root) => (
     root === '.' || file === root || file.startsWith(`${root}/`)
   ));

package/bin/delegate.cjs

@@ -12,6 +12,7 @@ const PERMISSION_SCHEMA = 'ultra-delegation-permission-v1';
 const RECEIPT_SCHEMA = 'ultra-delegation-receipt-v1';
 const RESULT_SCHEMA = 'ultra-delegation-result-v1';
 const PERMISSION_FIELDS = new Set(['$schema', 'writable_roots', 'external_effects']);
+const PROTECTED_ROOT = '.ultra';
 const RESULT_JSON_SCHEMA = Object.freeze({
   type: 'object',
   additionalProperties: false,
@@ -149,6 +150,12 @@ function readPermission(file, worktree) {
       throw new Error('permission writable_roots must not escape through symlinks');
     }
     const normalized = path.relative(worktree, real).split(path.sep).join('/') || '.';
+    // `.ultra` is the project's memory and has exactly one writer: the primary host.
+    // Rejected here so the delegation never starts, and again against the actual diff
+    // in the worker, which also covers the `.` grant.
+    if (normalized === PROTECTED_ROOT || normalized.startsWith(`${PROTECTED_ROOT}/`)) {
+      throw new Error(`permission writable_roots must not include ${PROTECTED_ROOT}; the primary host is its only writer`);
+    }
     if (!writableRoots.includes(normalized)) {
       writableRoots.push(normalized);
       writableRootPaths.push(real);

package/docs/ARCHITECTURE.md

@@ -38,14 +38,22 @@ The exact allowlists live in `adapters/_shared/runtime-assets.cjs`. Public workf
 never invoke another public workflow. They may use model-invoked disciplines and then
 recommend the next owner route.
 
+Research's seventeen references are workflow-owned semantic lenses. Its optional
+`wayfinding.md` maps a multi-lens question without becoming a new Skill or authority.
+The reusable mechanics remain separate: Grilling owns the interaction loop, Think owns
+one consequential decision, and Domain Modeling owns vocabulary promotion.
+
 ### Authority plane
 
 Canonical semantic authority consists of:
 
-- `.ultra/north-star.md` for owner wording and hard constraints;
+- `.ultra/project-brief.md` for raw owner intake and the initial outline;
+- `.ultra/north-star.md` for the accepted direction, outcome or metric decision, hard
+  constraints, and exclusions established by Research;
 - `.ultra/specs/*.md` for product, architecture, discovery, and research distillate;
-- `.ultra/changes/{active,archive}/<id>/intent.md` for bounded outcomes;
-- `.ultra/tasks.json` plus each task's `context_file`;
+- `.ultra/changes/{active,archive,abandoned}/<id>/intent.md` for bounded outcomes with
+  one stable id across directory moves;
+- append-only `.ultra/tasks.json` rows carrying `change_id`, plus each task's `context_file`;
 - `.ultra/decisions/*.md`, `.ultra/evidence/`, `.ultra/test-report.json`, and research;
 - repository-root `CONTEXT.md` for ubiquitous language;
 - Git for revision identity, history, moves, diff, and rollback.
@@ -53,6 +61,12 @@ Canonical semantic authority consists of:
 Each semantic fact has one canonical representation. A summary can point to authority,
 but cannot become a second authority merely because it is convenient to read.
 
+One primary host model writes this authority in a worktree. Native subagents are
+read-only sensors; delegated CLIs write isolated non-`.ultra` roots and return receipts.
+Sequential Changes share the ledger by stable id, while current readers filter by the
+one unique active Change. This keeps concurrency explicit in Git instead of adding a
+semantic lock service or workflow engine.
+
 ### Adaptation plane
 
 `adapters/*.js` translates the shared package into native host surfaces:
@@ -87,7 +101,10 @@ the parent model runs the same references sequentially.
 Five hooks read canonical files and emit bounded context or observations. Their output
 under `.ultra/.runtime/`, `.ultra/progress/`, or `.ultra/reviews/` is derived and
 disposable. The dangerous-command hook is the only hard effect guard; it classifies a
-small named set of shell effects and provides an exact-command authorization repair.
+small named destructive set and provides an exact-command authorization repair.
+Additive protected-branch publication remains advisory when portable host wiring cannot
+project a trusted owner-approval receipt; history rewrites and branch deletion remain
+guarded.
 
 ## Completion and recovery
 
@@ -97,10 +114,11 @@ end-to-end evidence by themselves.
 
 Every user Skill starts by reading:
 
-1. `.ultra/tasks.json`;
-2. the unfinished task's `context_file` and closing `## Resume Note`;
-3. `CONTEXT.md` and relevant decisions;
-4. the active Change, specifications, evidence, and current Git state.
+1. zero or one active `change_id`, diagnosing more than one;
+2. `.ultra/tasks.json` tasks whose `change_id` matches it;
+3. the matching frontier task's `context_file` and closing `## Resume Note`;
+4. `CONTEXT.md` and relevant decisions;
+5. the active Change, specifications, evidence, and current Git state.
 
 That same sequence works after compaction, a fresh process, disabled hooks, or a host
 change. Compact snapshots can accelerate recovery but never outrank the source files.
@@ -109,10 +127,12 @@ change. Compact snapshots can accelerate recovery but never outrank the source f
 
 Three independent Skills protect the vertical path:
 
-1. `ultra-plan` confirms seams and rejects horizontal feature tasks.
+1. `ultra-plan` records model-selected technical seams, asks the owner only for a
+   public-contract or material trade-off, and rejects horizontal feature tasks.
 2. `ultra-dev` records six separate evidence dimensions, including real persistence,
    default-on behavior, vertical execution, and spec trace.
-3. `ultra-test` searches exports for real consumers and exercises the primary flow.
+3. `ultra-test` searches exports for real consumers, exercises the primary flow, and
+   binds the result to Change id, task ids, intent, HEAD, and product-worktree digest.
 
 These are independent sensors. None rewrites a finding into product truth.
 
@@ -128,6 +148,12 @@ schema-constrained final response; the Node worker validates `finished`, `blocke
 writes its own receipt; stdout, stderr, native final output, and the launch receipt
 remain diagnostic files.
 
+The instruction binds the active `change_id` and one of three semantic scopes: task
+execution/continuation, scoped Research evidence, or aggregate Change
+review/verification. Only task execution requires a task row. The read-only Research
+and aggregate scopes preserve the same process boundary without inventing work merely
+to make delegation reachable.
+
 Delegation does not grant authority or copy semantic state into another store. The
 worker reads and writes the same repository files in the specified worktree.