@orkestrel/scaffold 0.0.108 → 0.0.110

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (162) hide show
  1. package/dist/agents/skills/orkestrel-dispatch/scripts/bench.js +326 -91
  2. package/dist/agents/skills/orkestrel-dispatch/scripts/brief.js +110 -14
  3. package/dist/agents/skills/orkestrel-dispatch/scripts/chain.js +280 -0
  4. package/dist/agents/skills/orkestrel-dispatch/scripts/cite.js +188 -26
  5. package/dist/agents/skills/orkestrel-dispatch/scripts/launch.js +142 -43
  6. package/dist/agents/skills/orkestrel-dispatch/scripts/login.js +240 -62
  7. package/dist/agents/skills/orkestrel-dispatch/scripts/patch.js +323 -0
  8. package/dist/agents/skills/orkestrel-dispatch/scripts/result.js +169 -32
  9. package/dist/agents/skills/orkestrel-dispatch/scripts/sweep.js +290 -69
  10. package/dist/agents/skills/orkestrel-falsify/scripts/mutate.js +560 -0
  11. package/dist/agents/skills/orkestrel-harden/scripts/discovery.js +1052 -205
  12. package/dist/agents/skills/orkestrel-harden/scripts/measure.js +361 -0
  13. package/dist/agents/skills/orkestrel-harden/scripts/titles.js +455 -80
  14. package/dist/agents/skills/orkestrel-parallel/scripts/cases.js +327 -0
  15. package/dist/agents/skills/orkestrel-publish/scripts/compare.js +298 -51
  16. package/dist/agents/skills/orkestrel-publish/scripts/pins.js +170 -13
  17. package/dist/agents/skills/orkestrel-publish/scripts/swap.js +956 -191
  18. package/dist/agents/skills/orkestrel-publish/scripts/wave.js +707 -112
  19. package/dist/agents/skills/orkestrel-publish/scripts/window.js +235 -41
  20. package/dist/agents/skills/orkestrel-scout/scripts/map.js +341 -54
  21. package/dist/agents/templates/brief.md +6 -2
  22. package/dist/bin/main.js +51 -46
  23. package/dist/bin/main.js.map +1 -1
  24. package/dist/host/AGENTS.md +23 -21
  25. package/dist/host/agents/orchestration.md +85 -54
  26. package/dist/host/agents/platforms/windows.md +8 -3
  27. package/dist/host/agents/skills/orkestrel-align/SKILL.md +10 -13
  28. package/dist/host/agents/skills/orkestrel-align/references/fleet.md +4 -10
  29. package/dist/host/agents/skills/orkestrel-align/references/integration.md +1 -1
  30. package/dist/host/agents/skills/orkestrel-build/SKILL.md +16 -48
  31. package/dist/host/agents/skills/orkestrel-build/agents/openai.yaml +1 -1
  32. package/dist/host/agents/skills/orkestrel-debrief/SKILL.md +19 -16
  33. package/dist/host/agents/skills/orkestrel-debrief/agents/openai.yaml +1 -1
  34. package/dist/host/agents/skills/orkestrel-debrief/references/field-testing.md +15 -28
  35. package/dist/host/agents/skills/orkestrel-debrief/references/instruction-audit.md +5 -6
  36. package/dist/host/agents/skills/orkestrel-debrief/references/retention.md +4 -6
  37. package/dist/host/agents/skills/orkestrel-dispatch/SKILL.md +36 -25
  38. package/dist/host/agents/skills/orkestrel-dispatch/references/launch.md +4 -5
  39. package/dist/host/agents/skills/orkestrel-dispatch/scripts/bench.ts +366 -75
  40. package/dist/host/agents/skills/orkestrel-dispatch/scripts/brief.ts +126 -16
  41. package/dist/host/agents/skills/orkestrel-dispatch/scripts/chain.ts +323 -0
  42. package/dist/host/agents/skills/orkestrel-dispatch/scripts/cite.ts +220 -31
  43. package/dist/host/agents/skills/orkestrel-dispatch/scripts/launch.ts +176 -33
  44. package/dist/host/agents/skills/orkestrel-dispatch/scripts/login.ts +294 -59
  45. package/dist/host/agents/skills/orkestrel-dispatch/scripts/patch.ts +401 -0
  46. package/dist/host/agents/skills/orkestrel-dispatch/scripts/result.ts +199 -34
  47. package/dist/host/agents/skills/orkestrel-dispatch/scripts/sweep.ts +330 -69
  48. package/dist/host/agents/skills/orkestrel-falsify/SKILL.md +14 -12
  49. package/dist/host/agents/skills/orkestrel-falsify/references/brief.md +2 -5
  50. package/dist/host/agents/skills/orkestrel-falsify/references/reconcile.md +2 -1
  51. package/dist/host/agents/skills/orkestrel-falsify/scripts/mutate.ts +701 -0
  52. package/dist/host/agents/skills/orkestrel-harden/SKILL.md +31 -24
  53. package/dist/host/agents/skills/orkestrel-harden/references/bindings.md +23 -81
  54. package/dist/host/agents/skills/orkestrel-harden/references/centralization.md +4 -5
  55. package/dist/host/agents/skills/orkestrel-harden/references/contract.md +2 -0
  56. package/dist/host/agents/skills/orkestrel-harden/references/document.md +33 -0
  57. package/dist/host/agents/skills/orkestrel-harden/references/equivalence.md +6 -3
  58. package/dist/host/agents/skills/orkestrel-harden/references/hardening.md +34 -18
  59. package/dist/host/agents/skills/orkestrel-harden/references/research.md +2 -0
  60. package/dist/host/agents/skills/orkestrel-harden/scripts/discovery.ts +1272 -279
  61. package/dist/host/agents/skills/orkestrel-harden/scripts/measure.ts +425 -0
  62. package/dist/host/agents/skills/orkestrel-harden/scripts/titles.ts +603 -98
  63. package/dist/host/agents/skills/orkestrel-journey/SKILL.md +79 -98
  64. package/dist/host/agents/skills/orkestrel-journey/references/captures.md +54 -17
  65. package/dist/host/agents/skills/orkestrel-journey/references/decide.md +1 -20
  66. package/dist/host/agents/skills/orkestrel-journey/references/layer.md +21 -31
  67. package/dist/host/agents/skills/orkestrel-journey/references/styles.md +1 -1
  68. package/dist/host/agents/skills/orkestrel-parallel/SKILL.md +33 -29
  69. package/dist/host/agents/skills/orkestrel-parallel/references/claude.md +15 -13
  70. package/dist/host/agents/skills/orkestrel-parallel/references/cloud.md +28 -0
  71. package/dist/host/agents/skills/orkestrel-parallel/references/codex.md +1 -0
  72. package/dist/host/agents/skills/orkestrel-parallel/references/cursor.md +13 -13
  73. package/dist/host/agents/skills/orkestrel-parallel/references/t3.md +13 -13
  74. package/dist/host/agents/skills/orkestrel-parallel/scripts/cases.ts +389 -0
  75. package/dist/host/agents/skills/orkestrel-polish/SKILL.md +16 -20
  76. package/dist/host/agents/skills/orkestrel-polish/references/capture-harness.md +16 -26
  77. package/dist/host/agents/skills/orkestrel-publish/SKILL.md +12 -26
  78. package/dist/host/agents/skills/orkestrel-publish/agents/openai.yaml +2 -2
  79. package/dist/host/agents/skills/orkestrel-publish/references/adopt.md +23 -0
  80. package/dist/host/agents/skills/orkestrel-publish/references/release.md +4 -4
  81. package/dist/host/agents/skills/orkestrel-publish/references/wave.md +13 -10
  82. package/dist/host/agents/skills/orkestrel-publish/references/window.md +21 -26
  83. package/dist/host/agents/skills/orkestrel-publish/scripts/compare.ts +348 -59
  84. package/dist/host/agents/skills/orkestrel-publish/scripts/pins.ts +202 -22
  85. package/dist/host/agents/skills/orkestrel-publish/scripts/swap.ts +1061 -191
  86. package/dist/host/agents/skills/orkestrel-publish/scripts/wave.ts +812 -140
  87. package/dist/host/agents/skills/orkestrel-publish/scripts/window.ts +280 -38
  88. package/dist/host/agents/skills/orkestrel-scout/SKILL.md +3 -3
  89. package/dist/host/agents/skills/orkestrel-scout/scripts/map.ts +428 -70
  90. package/dist/host/agents/templates/brief.md +6 -2
  91. package/dist/host/agents/transports/claude-cli.md +3 -3
  92. package/dist/host/agents/transports/codex.md +3 -8
  93. package/dist/host/agents/transports/cursor.md +2 -3
  94. package/dist/host/claude/AGENTS.md +2 -3
  95. package/dist/host/claude/agents/analyst.md +3 -3
  96. package/dist/host/claude/agents/astra.md +4 -4
  97. package/dist/host/claude/agents/builder.md +4 -3
  98. package/dist/host/claude/agents/checker.md +1 -1
  99. package/dist/host/claude/agents/distiller.md +1 -1
  100. package/dist/host/claude/agents/grok.md +1 -1
  101. package/dist/host/claude/agents/opus.md +3 -3
  102. package/dist/host/claude/agents/orkestrel.md +57 -90
  103. package/dist/host/claude/agents/researcher.md +1 -1
  104. package/dist/host/claude/agents/reviewer.md +6 -2
  105. package/dist/host/claude/agents/scout.md +1 -1
  106. package/dist/host/claude/agents/verifier.md +2 -2
  107. package/dist/host/claude/rules/application.md +14 -21
  108. package/dist/host/claude/rules/architecture.md +34 -44
  109. package/dist/host/claude/rules/browser.md +3 -4
  110. package/dist/host/claude/rules/campaign.md +3 -2
  111. package/dist/host/claude/rules/documentation.md +25 -12
  112. package/dist/host/claude/rules/names.md +9 -4
  113. package/dist/host/claude/rules/patterns.md +6 -7
  114. package/dist/host/claude/rules/portability.md +1 -1
  115. package/dist/host/claude/rules/quality.md +7 -6
  116. package/dist/host/claude/rules/styles.md +2 -2
  117. package/dist/host/claude/rules/tests.md +68 -115
  118. package/dist/host/claude/rules/typescript.md +7 -9
  119. package/dist/host/claude/rules/workspace.md +27 -26
  120. package/dist/host/claude/rules/writing.md +2 -5
  121. package/dist/host/claude/skills/orkestrel-dispatch/SKILL.md +1 -1
  122. package/dist/host/claude/skills/orkestrel-harden/SKILL.md +1 -1
  123. package/dist/host/claude/skills/orkestrel-publish/SKILL.md +1 -1
  124. package/dist/host/claude/skills/orkestrel-scout/SKILL.md +1 -1
  125. package/dist/host/codex/agents/analyst.toml +9 -3
  126. package/dist/host/codex/agents/astra.toml +10 -7
  127. package/dist/host/codex/agents/builder.toml +4 -4
  128. package/dist/host/codex/agents/opus.toml +6 -6
  129. package/dist/host/codex/agents/orkestrel.toml +1 -1
  130. package/dist/host/codex/agents/researcher.toml +2 -2
  131. package/dist/host/codex/agents/reviewer.toml +2 -2
  132. package/dist/host/codex/agents/verifier.toml +4 -4
  133. package/dist/host/codex/config.toml +1 -1
  134. package/dist/host/configs/helpers.ts +21 -1
  135. package/dist/host/configs/policy.ts +299 -15
  136. package/dist/host/cursor/rules/orchestration.mdc +1 -1
  137. package/dist/host/dotfiles/oxlintrc.json +137 -3
  138. package/dist/host/guides/agent.md +111 -134
  139. package/dist/host/guides/capture.md +801 -0
  140. package/dist/host/guides/probe.md +27 -8
  141. package/dist/host/guides/scaffold.md +97 -19
  142. package/dist/host/guides/study.md +534 -440
  143. package/dist/host/guides/test.md +111 -44
  144. package/dist/host/manifest.json +170 -244
  145. package/dist/host/tests/config.test.ts +977 -187
  146. package/dist/host/tests/policy.test.ts +165 -1
  147. package/dist/host/tests/setupPolicy.ts +1246 -33
  148. package/dist/src/core/index.cjs +222 -90
  149. package/dist/src/core/index.cjs.map +1 -1
  150. package/dist/src/core/index.d.cts +126 -40
  151. package/dist/src/core/index.d.ts +126 -40
  152. package/dist/src/core/index.js +217 -89
  153. package/dist/src/core/index.js.map +1 -1
  154. package/dist/src/server/index.cjs +20 -27
  155. package/dist/src/server/index.cjs.map +1 -1
  156. package/dist/src/server/index.js +21 -28
  157. package/dist/src/server/index.js.map +1 -1
  158. package/package.json +11 -11
  159. package/dist/agents/skills/orkestrel-dispatch/scripts/helpers.js +0 -283
  160. package/dist/host/agents/skills/orkestrel-dispatch/references/bench.md +0 -25
  161. package/dist/host/agents/skills/orkestrel-dispatch/scripts/helpers.ts +0 -333
  162. package/dist/host/guides/supervisor.md +0 -4890
@@ -9,7 +9,7 @@
9
9
  - `*/types.ts` is authoritative for public APIs. Never undo a user's type edit.
10
10
  - `.agents/orchestration.md` governs agent operation and cannot weaken this file. Harness bridges (`.claude/AGENTS.md`, `.codex/config.toml`, `.cursor/rules/orchestration.mdc`) add harness mechanics only.
11
11
  - A skill under `.agents/skills/` is a procedure. Follow it when the user or a dispatch names it, or when its trigger fires. A skill cannot weaken this file or a rule.
12
- - Before editing, read: this file; each rule the Rule map scopes to the paths you touch; the named skill and the references it names; the matching guide under `guides/` and `ROADMAP.md` when present. Read nothing else up front.
12
+ - Before editing, read: this file; each rule whose `paths` frontmatter matches the paths you touch; the named skill and the references it names; the matching guide under `guides/` and `ROADMAP.md` when present. Read nothing else up front.
13
13
  - Rules state how to write. Guides state what to build. On conflict, stop and report the conflict.
14
14
  - A rule file's stated exception to a law in this file applies.
15
15
 
@@ -43,7 +43,7 @@ configs/ thin target wrappers around root Vite/TypeScript configuration
43
43
  - **NEVER** use mocks, behavioral fakes, module replacement, framework spies, or fake clocks for project-owned behavior. Use real implementations, recorders, temporary resources, protocol-faithful fixture servers, and inert data stubs.
44
44
  - **ALWAYS** make interface properties and public return collections readonly.
45
45
  - **ALWAYS** define reusable and public types in `*/types.ts` before implementation.
46
- - **ALWAYS** inspect the declared and installed `@orkestrel/*` capabilities before writing overlapping logic. Reuse a primitive whose semantics match; never wrap one to rename it.
46
+ - **ALWAYS** check whether an `@orkestrel` package already provides a capability before writing logic, a tool, or an instrument that could overlap it. Read the package guides first: `guides/README.md` maps them, and a target holds them under `node_modules/@orkestrel/scaffold/dist/host/guides/`. When a guide leaves the question open, read the package's source in a checkout beside the repository. When no checkout exists, read its repository at `https://github.com/orkestrel/<package>`, online or cloned under `tmp/`. Reuse a primitive whose semantics match; never wrap one to rename it.
47
47
  - **ALWAYS** finish the requested implementation: no stubs, deferred logic, or hidden follow-up.
48
48
  - **ALWAYS** write a script as TypeScript run by Node (`node path/to/script.ts`, type stripping, `node:` modules only): a skill script, a probe, a launcher, an instrument, a one-off tool. Never write a bash, PowerShell, or Python script. The Claude Code Cloud hooks under `scripts/` are the one bash exception, and that folder holds nothing else.
49
49
 
@@ -72,6 +72,8 @@ configs/ thin target wrappers around root Vite/TypeScript configuration
72
72
 
73
73
  Types, tests, and implementation decide every step; comments, TSDoc, and guides follow the final implementation. Report results against the goal, and prefer the smallest change that reaches green over a perfect one.
74
74
 
75
+ Open a task with its goal, its acceptance criteria, and the cheapest pass/fail check that proves them. Before writing a fixture, proof, or tool around a command, run that command on real data. `.agents/orchestration.md` § Depth, breadth, and machinery sizes the effort and rules on what becomes a package.
76
+
75
77
  Size the change before touching a file. Take the largest row whose trigger applies: large, then medium, then small. Run only what the row names.
76
78
 
77
79
  | Size | Trigger | Do |
@@ -89,7 +91,7 @@ Within the size, in order:
89
91
  3. **Implement.** Conform to the types. Place each declaration in its kind file. Update the barrel.
90
92
  4. **Test.** Cover happy paths, edge cases, failures, and boundaries with deterministic tests against real implementations. For a defect, record the failing command and count before the fix and the same command green after.
91
93
  5. **Consolidate.** Remove duplication, nested functions, and superfluous wrappers without widening the API.
92
- 6. **Document.** Once the implementation is final, update the matching guide, comments, and TSDoc in one minimal pass. Before then, write, revise, and review no prose.
94
+ 6. **Document.** After the implementation is final, update the matching guide, comments, and TSDoc in one minimal pass. Before then, write, revise, and review no prose.
93
95
 
94
96
  Test ladder, narrowest first; stop at the boundary the change size names:
95
97
 
@@ -108,24 +110,24 @@ Test ladder, narrowest first; stop at the boundary the change size names:
108
110
 
109
111
  ## Rule map
110
112
 
111
- Each row is a normative extension of this file. Its `paths` frontmatter controls automatic loading; read the row when its subject matches your change.
112
-
113
- | Rule | Governs |
114
- | -------------------------------- | ---------------------------------------------------------------------- |
115
- | `.claude/rules/names.md` | Identifiers, API shape, acronyms, lifecycle vocabulary, files/folders |
116
- | `.claude/rules/typescript.md` | TypeScript syntax, imports, immutability, errors, TSDoc |
117
- | `.claude/rules/architecture.md` | Centralized files, exports, classes, modules, extension points, stores |
118
- | `.claude/rules/patterns.md` | Options, managers, emitters, guards, parsers, contracts |
119
- | `.claude/rules/tests.md` | Test behavior, helpers, probes, browser tests, Vitest configuration |
120
- | `.claude/rules/workspace.md` | Src/app environments, aliases, isolation, builds, scripts, tooling |
121
- | `.claude/rules/application.md` | App composition, entries, manifest safety, lifecycle, integration |
122
- | `.claude/rules/browser.md` | Vue/browser architecture and platform usage |
123
- | `.claude/rules/styles.md` | SCSS/CSS centralization, tokens, mixins, layers, naming |
124
- | `.claude/rules/portability.md` | Host branching, paths, line endings, processes, terminals, OS claims |
125
- | `.claude/rules/documentation.md` | Guides, parity, roadmap, showcase, examples, skill files |
126
- | `.claude/rules/writing.md` | Prose in guides, replies, commits, and instruction files |
127
- | `.claude/rules/quality.md` | Evidence, probes, instruments, performance, research, completion |
128
- | `.claude/rules/campaign.md` | The `.orkestrel/` campaign folder: layout, content, session start |
113
+ Each row is a normative extension of this file. A rule's `paths` frontmatter controls automatic loading, and `.claude/rules/writing.md` carries none and loads always. Read the row when its subject matches your change.
114
+
115
+ | Rule | Governs |
116
+ | -------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- |
117
+ | `.claude/rules/names.md` | Identifiers, API shape, helper prefixes, acronyms, fleet name ownership, lifecycle vocabulary, files and folders |
118
+ | `.claude/rules/typescript.md` | TypeScript syntax, imports, types, immutability, bindings and state, errors, TSDoc |
119
+ | `.claude/rules/architecture.md` | Centralized files, declaration placement, kind purity, wrappers, functions, classes, middleware, plugins, stores, barrels |
120
+ | `.claude/rules/patterns.md` | Ecosystem reuse, options, managers, emitters, validation and contracts, foreign contracts |
121
+ | `.claude/rules/tests.md` | Test contract, cross-cutting proofs, probes, live-service and expensive proofs, shared helpers, browser tests, runner configuration |
122
+ | `.claude/rules/workspace.md` | Src/app environments, aliases, configuration, builds, test project matrix, typechecking scopes, scripts, policy instruments, text integrity |
123
+ | `.claude/rules/application.md` | App composition, CLI selections, entries, manifests, server runner |
124
+ | `.claude/rules/browser.md` | Vue and browser architecture: composables, reactivity, platform APIs |
125
+ | `.claude/rules/styles.md` | SCSS/CSS centralization, tokens, mixins, layers, folders, proofs, naming |
126
+ | `.claude/rules/portability.md` | Host branching, line endings, paths, processes, terminals, scripts and packaging, host claims |
127
+ | `.claude/rules/documentation.md` | Guides, parity, examples, agent-facing surfaces, skill files |
128
+ | `.claude/rules/writing.md` | Prose in guides, replies, commits, and instruction files |
129
+ | `.claude/rules/quality.md` | Evidence, probes, instruments, performance, ecosystem reuse, hardening, completion |
130
+ | `.claude/rules/campaign.md` | The `.orkestrel/` campaign folder: layout, content, session start, hand off and close |
129
131
 
130
132
  ## Writing
131
133
 
@@ -12,47 +12,50 @@ On a harness with no bridge, orchestrate natively, record the harness in the ros
12
12
 
13
13
  Name the harness family and its access route: Claude through Claude Code or the Claude CLI; Codex through Codex native agents or the Codex CLI; Cursor through the Cursor CLI, with Grok 4.7 Extra High for the Grok bench. Classify GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna as Codex-family engines, including when another harness hosts the Orchestrator.
14
14
 
15
- | Engine | Job | Posture |
16
- | ----------------------- | ------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
17
- | Claude Fable 5.1 | orchestration in Claude Code: the plan, every decision, integration, acceptance | never dispatched: never a subagent, a bench, a fork, or a Workflow node, except as the consultant the user names for a wording and coherence pass after research and technical review |
18
- | Cursor Grok | context absorption, distillation, broad mapping, exploration | read-only; returns evidence, never decisions |
19
- | Claude Opus 5.5 | subjective design, design-fit review, implementation | proposes, audits, implements; never accepts its own work |
20
- | GPT-6 Astra | objective analysis, correctness audit, implementation | proposes, audits, implements; never accepts its own work |
21
- | Claude Sonnet 5.5 | bounded implementation of a closed shape, ecosystem reconciliation | executes a fully specified brief |
22
- | Claude Haiku 5.5 | bench drivers, small precise tasks, exact lookups, conformance, gates | executes a fully specified brief; edits no source |
23
- | GPT-6.1 Sol, GPT-6 Luna | mechanical units, drivers, small precise tasks, gates, exact lookups | executes a fully specified brief |
15
+ | Engine | Job | Posture |
16
+ | ----------------- | ------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
17
+ | Claude Fable 5.1 | orchestration in Claude Code: the plan, every decision, integration, acceptance | never dispatched as a writer, a bench, a fork, or a Workflow node; dispatched only as a read-only consultant, for a bounded design question at the stop the Orchestrator set for a seam, or for the wording and coherence pass the user names |
18
+ | Cursor Grok | context absorption, distillation, broad mapping, exploration | as a bench: read-only; returns evidence, never decisions |
19
+ | Claude Opus 5.5 | subjective design, design-fit review, implementation | proposes, audits, implements; never accepts its own work |
20
+ | GPT-6 Astra | objective analysis, correctness audit, implementation | proposes, audits, implements; never accepts its own work |
21
+ | Claude Sonnet 5.5 | bounded implementation of a closed shape, ecosystem reconciliation | executes a fully specified brief |
22
+ | Claude Haiku 5.5 | bench drivers, small precise tasks, exact lookups, conformance, gates | executes a fully specified brief; edits no source |
23
+ | GPT-6.1 Sol | a fully specified unit with a closed shape | executes a fully specified brief |
24
+ | GPT-6 Luna | mechanical units, drivers, small precise tasks, gates, exact lookups | executes a fully specified brief |
24
25
 
25
26
  The harness's own engine orchestrates: Claude Fable 5.1 or Claude Opus 5.5 in Claude Code, GPT-6 Astra in Codex, Cursor Grok in Cursor. The Orchestrator owns the plan, every decision, integration, and acceptance. When the harness is Codex or the Orchestrator uses a Codex-family engine, use the Codex column: map Opus to GPT-6 Astra, Sonnet to GPT-6.1 Sol, and Haiku to GPT-6 Luna; execute mapped Astra roles natively without an Opus bench probe. Under this mapping, assign review to another agent in a clean context and forbid self-review, including when writer and reviewer both use Astra. For other harnesses, review runs on an engine that did not write the work.
26
27
 
27
28
  ## Roles
28
29
 
29
- | Role | Job | Capability | Claude | Codex |
30
- | ------------ | ---------------------------------------------------------- | ------------ | ------------ | ----------- |
31
- | `grok` | absorption, distillation, broad mapping, exploration | reader | haiku driver | luna driver |
32
- | `analyst` | objective design argument or correctness audit on Astra | objective | haiku driver | astra |
33
- | `astra` | objective, constraint-heavy implementation on Astra | objective | haiku driver | astra |
34
- | `planner` | subjective design | subjective | opus | astra |
35
- | `reviewer` | design-fit or correctness review | subjective | opus | astra |
36
- | `opus` | subjective implementation (API shape, naming, guide voice) | subjective | opus | astra |
37
- | `builder` | one bounded unit with a closed shape, app layer included | closed-shape | sonnet | sol |
38
- | `checker` | mechanical conformance against criteria and rules | mechanical | haiku | luna |
39
- | `verifier` | gates and evidence commands, exit-code truth | mechanical | haiku | luna |
40
- | `scout` | exact bounded lookup of files, symbols, seams | mechanical | haiku | luna |
41
- | `distiller` | bulk reading when the Cursor bench is dark | reader | haiku | luna |
42
- | `researcher` | primary-source research when the Cursor bench is dark | reader | haiku | luna |
43
- | `orkestrel` | ecosystem reconciliation over supplied evidence | closed-shape | sonnet | sol |
44
-
45
- - Name the role and its engine in every dispatch. Reach a role by its own name.
30
+ | Role | Job | Capability | Claude | Codex |
31
+ | ------------ | -------------------------------------------------------------- | ------------ | ------------ | ----------- |
32
+ | `grok` | absorption, distillation, broad mapping, exploration | reader | haiku driver | luna driver |
33
+ | `analyst` | objective design argument or correctness audit on Astra | objective | haiku driver | astra |
34
+ | `astra` | objective, constraint-heavy implementation on Astra | objective | haiku driver | astra |
35
+ | `planner` | subjective design | subjective | opus | astra |
36
+ | `reviewer` | design-fit or correctness review | subjective | opus | astra |
37
+ | `opus` | subjective implementation (API shape, naming, guide voice) | subjective | opus | astra |
38
+ | `builder` | a fully specified unit with a closed shape, app layer included | closed-shape | sonnet | sol |
39
+ | `checker` | mechanical conformance against criteria and rules | mechanical | haiku | luna |
40
+ | `verifier` | gates and evidence commands, exit-code truth | mechanical | haiku | luna |
41
+ | `scout` | exact bounded lookup of files, symbols, seams | mechanical | haiku | luna |
42
+ | `distiller` | bulk reading when the Cursor bench is dark | reader | haiku | luna |
43
+ | `researcher` | primary-source research when the Cursor bench is dark | reader | haiku | luna |
44
+ | `orkestrel` | ecosystem reconciliation over supplied evidence | reader | sonnet | sol |
45
+
46
+ - Name the role and its engine in every dispatch. Reach a role by its own name. When a hosting harness offers both columns, the dispatch names the column each lane runs on.
47
+ - The capability column names the lane a role holds by default and the one a dispatch may assign (`reviewer`: subjective by default, objective when assigned; `orkestrel`: reader).
46
48
  - On a harness with no column, map each capability to an engine in the roster and route by that map; add the harness's column here when it carries campaign work.
47
49
  - A driver launches another harness's CLI and returns the journal path and session id with the result. It never judges or implements. Refuse a bench result that carries no journal path and session id; it ran on the driver.
48
50
  - `.agents/transports/<harness>.md` owns each bench's invocation, sandbox, journal, and recovery. `.claude/agents/` and `.codex/agents/` hold one file per role.
49
51
 
50
52
  ## Routing
51
53
 
52
- - Route by judgment load. Objective, constraint-heavy, mechanical-precision work goes to `astra`. API shape, naming, and documentation voice go to `opus`. A bounded unit whose brief or types fix its names, API, and architecture goes to `builder`, whatever its private lines; a unit that spans packages or leaves one of those open goes to `astra` or `opus`.
54
+ - Route by judgment load. Objective, constraint-heavy, mechanical-precision work goes to `astra`. API shape, naming, and documentation voice go to `opus`. A bounded unit whose brief or types fix its names, API, and architecture goes to `builder`; a unit that spans packages or leaves one of those open goes to `astra` or `opus`. Route a unit that creates or changes a cross-process or filesystem protocol (a lock, a marker, an atomic publish, termination handling, or a retry class) to `astra` or `opus`; a unit that leaves its mechanism open is not closed-shape, however closed its public shape looks.
53
55
  - Route all context absorption, distillation, broad mapping, and exploratory or fishing reads to `grok` on Cursor Grok 4.7 Extra High: terrain, prior art, diff sweeps, scattered sources. Keep decision-bearing reading with the decision owner. Fall back to `distiller` or `researcher` only when the Cursor bench is dark, and record the fallback; ease or cost never authorizes a fallback.
54
56
  - Keep small tedious precise tasks native: Claude Haiku in Claude, Codex Luna in Codex. Use `scout` for an exact bounded lookup; keep bench drivers, `checker`, and `verifier` on their mapped native engines. A bounded file scope does not turn context absorption or exploratory reading into a native lookup.
55
- - Route gates to `verifier`. A writer's self-reported gate is not evidence.
57
+ - Route a unit that authors path-containment, import-resolution-boundary, unescaped-input, or credential-exposure proofs to `opus` from the start where `opus` runs on Claude, and record the Codex bench dark for that unit only. Under native Codex roles, report a harness refusal through § Deviation protocol.
58
+ - Route gates to `verifier`. A writer's self-reported gate is not evidence. A writer never holds the runner lane for its own acceptance; a gate a writer runs is an observation, never acceptance.
56
59
  - Under the Codex mapping, put falsify lanes on Astra or Sol at high effort; keep Luna for gates and verifier runs, never for a lane that must resolve runtime claims.
57
60
  - Work directly on a one-file change or a one-line fix, subject to the preceding routing rules. Route standalone lookups through `scout`. Dispatch when isolation, parallelism, a second engine, or a large read pays for the brief.
58
61
 
@@ -63,8 +66,8 @@ The harness's own engine orchestrates: Claude Fable 5.1 or Claude Opus 5.5 in Cl
63
66
  - Record the session identity separately from its task, turn, process, and message address. Record each session's harness, provider, model, options, workspace, responsibility, owned paths, continuation address, most recent task, and evidence paths in the roster, and preserve its prior task results. The roster is `coord/roster.md` of the live `orkestrel-parallel` round, or `tmp/units/roster.md` when no round is live.
64
67
  - Treat a live writer as an agent with an active assignment and edit authority. A retained session without an assignment holds no resource merely by existing. Preserve an explicit hold until its owner releases it.
65
68
  - Continue a unit through compaction. Before a handoff or context limit, record decisions, changed instructions, source identity, unresolved questions, failed and passing evidence, active process identities, ownership, and the next action. Compaction neither completes a task nor releases its files or processes.
66
- - On continuation, the Orchestrator sends the next bounded objective and the facts changed after the last turn. The continued agent reads changed or missing authority and the source regions that objective needs, and rechecks mutable facts such as source state and process liveness. Do not repeat unchanged bulk reads merely because a unit resumes.
67
- - Consult a suitable peer when a consequential approach remains open, evidence conflicts, or repeated attempts stop narrowing a failure. Give the peer the question, relevant evidence, attempted approaches, constraints, and requested answer. Use an existing specialist when its context helps and a fresh peer when the question needs an independent challenge.
69
+ - On continuation, the Orchestrator sends the next bounded objective and the facts changed after the last turn. The continued agent reads changed or missing authority and the source regions that objective needs, and rechecks mutable facts such as source state and process liveness. Do not repeat unchanged bulk reads merely because a unit resumes. A continuation that changes scope or acceptance, or authorizes a fix, is written first as `<unit>-brief-<n>.md` and sent as a pointer to that file; a message alone continues only an unchanged objective.
70
+ - Consult a suitable peer at the stop set for a seam, or earlier when a consequential approach remains open, evidence conflicts, or attempts stop narrowing a failure, without waiting for the user, and record the question, recommendation, evidence, and next check in the plan. Give the peer the question, relevant evidence, attempted approaches, constraints, and requested answer. Use an existing specialist when its context helps and a fresh peer when the question needs an independent challenge.
68
71
  - Keep consultation read-only and bounded. Ask for a recommendation, its evidence, uncertainty, and a concrete next check. The executor retains the work; the Orchestrator rules on scope and acceptance. Advice grants no authority and does not satisfy an independent review gate.
69
72
  - Resolve conflicting advice against the contract and observable evidence. Do not add consultation to routine steps or repeat it without a changed question or evidence.
70
73
  - When a requested launch channel fails, record the failure and whether a child was created. Before retrying an uncertain launch, inspect the recorded task or session. Use a permitted fallback with its actual model and options disclosed; retry a refused route only after a relevant condition changes.
@@ -77,10 +80,33 @@ The harness's own engine orchestrates: Claude Fable 5.1 or Claude Opus 5.5 in Cl
77
80
  ## Pace
78
81
 
79
82
  - Spend process only where it changes a result. Launch, run, and report: no fixed clock windows, readiness-only turns, per-command acknowledgments, hash pins of a brief, or process queries beyond the recorded process ids.
80
- - Re-run a check only when its inputs changed; never repeat a passing check to collect another receipt.
81
- - Ask the user only for a decision that is theirs: scope, a conflicting ruling, a credential, publication, or a commit or push the law reserves. Decide routine matters yourself, such as time and lane allowances, retries, and order, and report them.
83
+ - When a verifier lane fails in its harness rather than in the gate, run the gate once yourself and read its exit code; write no comparator, snapshot, or retry chain to obtain evidence.
84
+ - Keep valid receipts. Re-run a check only when its inputs, configuration, environment, or claim changed; never repeat a passing check to collect another receipt.
85
+ - Ask the user only for a decision that is theirs: scope, the spend ceiling and stop line § Hold the plan names, a conflicting ruling, a credential, publication, or a commit or push the law reserves. Decide routine matters yourself, such as time and lane allowances, retries, and order, and report them, with time in whole minutes.
82
86
  - Never end a turn with work pending and nothing that will wake the session: run the work as a tracked background command or a child whose completion notifies, or schedule a wake. While a lane runs, continue independent work.
83
- - Keep the goal, the current step, and the next action visible in the campaign plan, and report progress against the goal at each milestone.
87
+
88
+ ## Hold the plan
89
+
90
+ - Open with the goal and acceptance criteria per `AGENTS.md` § Work loop, then estimate the scope: the units, the review and evidence each needs, and what stays out. When the estimate is large, agree with the user the estimate, the spend ceiling, and the stop line: how much lane time the work can take, and the point at which work stops and reports. Derive an estimate or a process cap from a measurement of the same command, execution shape, and toolchain, and state that measurement beside the figure, or state that none exists.
91
+ - Before each unit, name the plan step it serves. Cut a unit no step needs; set an unplanned idea aside in the plan and return to the step.
92
+ - Add a tracked file only where the request, the plan, or a rule names it.
93
+ - After every unit return, unexpected result, or user message, compare the work with the plan: mark what is done, restate the next step, and set aside what no longer serves the goal.
94
+ - When the work outgrows its estimate, stop, re-estimate, and bring the user the choice to cut, defer, or extend before continuing.
95
+ - When the user questions the plan, test it against the evidence in hand: reopen it where the evidence disagrees and report what changed; where the evidence holds, answer with that evidence.
96
+ - Answer a status question before dispatching more work, and close every report the same way: the goal, what is done against it, what is measured, what remains, and the next step.
97
+
98
+ ## Depth, breadth, and machinery
99
+
100
+ - Size the effort to the task before spending: a lookup takes one lane, routed per § Routing, and only open-ended work with independent, high-value parts fans out, per § Parallelism.
101
+ - Locate the evidence with short broad probes before any deep pass: a run of the shipped command, a census, a search. Go deep only on a defect or gap a run has shown, in one lane, with the narrowest instrument that answers it: a counting probe or a mutation.
102
+ - When a failure recurs, read the exact input and output of the failing case before adding capability, and fix the cause.
103
+ - Deepen in rounds and report a usable result after each. Stop at the first result that clears the acceptance criteria and name the gaps that remain.
104
+ - Prove a change in one consumer, with its gate green, before spreading it to the rest.
105
+ - Use the simplest method that can clear the criteria, starting with what an `@orkestrel` package already provides, looked up per `AGENTS.md` § Non-negotiable rules. Add machinery only after that method fails a measured check, and build nothing for a hypothetical need.
106
+ - Do a one-use step inline. Keep a tool one unit runs more than once as a throwaway under `tmp/`, deleted with the unit and never committed.
107
+ - When a second unit, package, or campaign needs a tool, move it out of `tmp/`: into a skill script under `.agents/skills/` when it only runs a procedure, and into its own package, with its own tests and release, when it carries correctness claims of its own or an identity consumers pin, replay, or audit. Write the promotion path and the removal path when you first build it.
108
+ - When a throwaway grows a mechanism per step instead of answering its question, stop and map it against the owning packages before building more.
109
+ - Ship the proven fix first, and run experiments behind it under a stated cap.
84
110
 
85
111
  ## Size gate
86
112
 
@@ -92,10 +118,13 @@ The harness's own engine orchestrates: Claude Fable 5.1 or Claude Opus 5.5 in Cl
92
118
  | medium | one opinion (`planner` or `analyst`) only when the shape is open | `opus`, `astra`, or `builder` by judgment load | one pass by an independent reviewer per § Engines, on numbered claims about the contract and the risky seams | `verifier` on the touched projects |
93
119
  | large | one independent round: `planner` and `analyst`, same brief, clean contexts, blind to each other | disjoint units in parallel | one `orkestrel-falsify` round on the integrated result | `verifier` tree-wide, once |
94
120
 
95
- - Never run a review on a small change. Never run a second design round on the same brief.
121
+ - Never run a review on a small change. Never run a second design round on the same brief. A design question a repair round opens is its own brief, not a second design round on the campaign's brief.
96
122
  - A review judges types, tests, and implementation. It reports comment, TSDoc, or guide wording only where the text contradicts the final behavior, and no fix round opens for prose alone.
97
- - A fix round re-audits the repaired claim only. A fix that adopted the reviewer's prescription verbatim closes with a mutation probe (disable the load-bearing line, watch the test fail, restore it). A fix that departs from it gets one pass by an independent reviewer per § Engines.
98
- - When review rounds at one seam keep surfacing defects without converging on their source, follow `.agents/skills/orkestrel-falsify/SKILL.md` § End the depth search.
123
+ - Before a repair is dispatched, class its remedy as a correction, a design question, or a deferral; a remedy that adds a mechanism (a file operation, a state, a protocol step, or a retry class) is a design question that goes to one `planner` or `analyst` pass carrying measured platform facts.
124
+ - A fix round re-audits the repaired claim only. A fix that departs from the reviewer's prescription gets one pass by an independent reviewer per § Engines.
125
+ - Every repair, a prescription adopted verbatim included, closes with a mutation probe on its load-bearing line (disable it, watch the test fail, restore it); `node .agents/skills/orkestrel-falsify/scripts/mutate.ts --edits EDITS.md -- COMMAND` runs the probe and proves the restore. Where the repair adds or changes a process proof, the proof's repeat run under suite load closes it too.
126
+ - Before another round at the same seam, weigh whether a further repair beats a design question, set and state in the brief the stop for that seam and its reason, re-examine it at every round, and end the depth search with a design question when a round introduces as many defects as it closes, a repair adds a mechanism, findings stop converging, or the cost outgrows the goal's importance. `.agents/skills/orkestrel-falsify/SKILL.md` § End the depth search owns how the search ends.
127
+ - A repair round at a seam carries only rows whose remedy serves that seam; a finding outside it waits for the next round or the deferral list.
99
128
  - An all-confirmed round ends the audit. A round needs an added or repaired claim to test; reviewer appetite is not a subject.
100
129
 
101
130
  ## Campaign
@@ -105,7 +134,7 @@ A large change opens with a concise goal and its exit criterion, short enough to
105
134
  1. **Absorb.** Map the terrain with the `orkestrel-scout` skill's `map.ts`, then `grok` distills it and the prior art. In an Orkestrel repository, `verifier` or another tool-capable role collects manifests, lockfiles, installed declarations, and registry readings, and `orkestrel` reconciles them.
106
135
  2. **Design.** One independent round: `planner` (subjective) and `analyst` (objective) on one brief, clean contexts, blind to each other. Reconcile into one plan: units, owned files, dependencies, parallel and serial order, acceptance criteria, exit criterion, ledger.
107
136
  3. **Implement.** Dispatch units per § Routing and § Parallelism.
108
- 4. **Integrate.** Apply returned patches to shared files serially. A type, mechanism, or criterion found here is a successor unit, never an integration edit.
137
+ 4. **Integrate.** Apply returned patches to shared files serially with `node .agents/skills/orkestrel-dispatch/scripts/patch.ts EDITS.md --plan`, then `--apply`. A type, mechanism, or criterion found here is a successor unit, never an integration edit.
109
138
  5. **Audit.** One `orkestrel-falsify` round on the integrated result, an independent reviewer per § Engines in each lane; add `checker` when the criteria are mechanical.
110
139
  6. **Verify.** `verifier` runs the tree-wide gates once.
111
140
  7. **Re-baseline.** Strike satisfied units, restate transformed ones, add units the exit criterion already required. A re-baseline never moves the exit criterion; that needs the user.
@@ -116,11 +145,11 @@ A large change opens with a concise goal and its exit criterion, short enough to
116
145
  - Parallelize independently owned work when useful progress can overlap. Reuse suitable sessions per § Sessions and consultation, bound each objective, and count coordination and verification costs when choosing the lanes. § Cache governs model and tool settings.
117
146
  - Judgment stays narrow: a design or review pass is at most the two lanes the size gate names, each in a clean context, blind to the other, and it runs once. Never fan judgment out further, and never run it as a loop.
118
147
  - Keep every lane, in this session and in peer sessions on the same machine, within the host budget of `.agents/skills/orkestrel-parallel/SKILL.md` § Bound the host.
119
- - One writer per checkout, except under `orkestrel-parallel`: there, at most three writers share one checkout, each owns disjoint files and disjoint regions of shared files, and one runner lane owns every test, check, and build process. Otherwise give parallel writers disjoint owned files, and give a writer its own worktree under `tmp/worktrees/` only when two units must write the same files at once. Commit a checkpoint before a writing dispatch.
148
+ - One writer per checkout, except under `orkestrel-parallel`: there, as many writers share one checkout as the host budget and disjoint ownership allow, each owning disjoint files and regions, and one runner lane owns every test, build, and tree-wide gate; `.agents/skills/orkestrel-parallel/SKILL.md` § Implement in phases names the checks a writer runs on its owned files. Otherwise give parallel writers disjoint owned files, and give a writer its own worktree under `tmp/worktrees/` only when two units must write the same files at once. Commit a checkpoint before a writing dispatch.
120
149
  - Fold every branch, clone, copy, or worktree an agent creates back into the original checkout's main branch with its changes left uncommitted there, so the user sees the whole difference from that branch. Commit inside the branch, clone, copy, or worktree as its unit needs, then remove it and delete its branch.
121
150
  - In a checkout another session shares, read the branch and `git status` before a write, and stop and ask the user when another session's uncommitted work is present.
122
- - Shared files are report-only; a unit returns an exact patch. Under `orkestrel-parallel`, a writer edits its owned regions of a shared file directly.
123
- - Scope validation by the claim and its input dependencies, not by writer ownership alone. Under `orkestrel-parallel`, the runner owns checks and obtains the required write holds. Preserve valid receipts; rerun a check when its inputs, configuration, environment, or claim changed. Never run tree-wide `format`, lint `--fix`, or `build` beside a live writer.
151
+ - Shared files are report-only; a unit returns an exact patch as an edit file that `.agents/skills/orkestrel-dispatch/scripts/patch.ts` applies. Under `orkestrel-parallel`, a writer edits its owned regions of a shared file directly.
152
+ - Scope validation by the claim and its input dependencies, not by writer ownership alone. Under `orkestrel-parallel`, the runner owns every test, build, and tree-wide gate (§ Implement in phases names a writer's owned-file checks) and obtains the required write holds; § Pace rules on reruns. Never run tree-wide `format`, lint `--fix`, or `build` beside a live writer.
124
153
  - One Grok lane at a time; queue the rest. An empty lane beside a live probe is starvation, not darkness.
125
154
  - The Orchestrator's own sweep is a writing dispatch: run it before or after the units, never beside them.
126
155
 
@@ -140,15 +169,15 @@ A large change opens with a concise goal and its exit criterion, short enough to
140
169
 
141
170
  Every working file lives under the checkout's gitignored `tmp/`, in the directory its kind names.
142
171
 
143
- | Directory | Holds |
144
- | ------------------------------------------ | -------------------------------------------------------------------------------------------------- |
145
- | `tmp/claude/`, `tmp/codex/`, `tmp/cursor/` | bench briefs, launch scripts, journals, `.err` files, last-message files, login logs, per harness |
146
- | `tmp/units/` | native unit briefs, reports, claims files, and unit records |
147
- | `tmp/probes/` | runtime probes the `probe` Vitest project collects and the `probe` MCP server arms |
148
- | `tmp/type/` | the type stage's workspace mirror, owned by `@orkestrel/probe` |
149
- | `tmp/captures/` | screenshots, resolved-style snapshots, and other capture portfolios |
150
- | `tmp/browsers/` | the journeys a `browse` server saves, their runs and captures, and its profiles under `.profiles/` |
151
- | `tmp/worktrees/` | worktrees the Orchestrator creates for a parallel writer |
172
+ | Directory | Holds |
173
+ | ------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
174
+ | `tmp/claude/`, `tmp/codex/`, `tmp/cursor/` | bench briefs, launch scripts, journals, `.err` files, last-message files, login logs, per harness |
175
+ | `tmp/units/` | native unit briefs, reports, claims files, and unit records |
176
+ | `tmp/probes/` | runtime probes the `probe` Vitest project collects and the `probe` MCP server arms, and a skill script's scratch under its own subdirectory (`census/`, `lint/`, `mutate/`) |
177
+ | `tmp/type/` | the type stage's workspace mirror, owned by `@orkestrel/probe` |
178
+ | `tmp/captures/` | screenshots, resolved-style snapshots, and other capture portfolios |
179
+ | `tmp/browsers/` | the journeys a `browse` server saves, their runs and captures, and its profiles under `.profiles/` |
180
+ | `tmp/worktrees/` | worktrees the Orchestrator creates for a parallel writer |
152
181
 
153
182
  ## .orkestrel layout
154
183
 
@@ -158,7 +187,7 @@ Every working file lives under the checkout's gitignored `tmp/`, in the director
158
187
 
159
188
  `orkestrel-dispatch` owns the brief template, launch mechanics for native and bench units, long-running commands, liveness, and recovery. Load it before a bench lane, a campaign unit, or any command that outlives the turn. The floor that follows binds every dispatch.
160
189
 
161
- - A small native unit's brief is the dispatch message: scope, law references, expected result. Write a brief file with the dispatch skill's `scripts/brief.ts` at `tmp/units/<unit>-brief.md` (native) or `tmp/<bench>/<unit>-brief.md` (bench) when the unit is a bench lane, belongs to a campaign, or might need re-running. A re-run gets `<unit>-brief-<n>.md`; never edit a launched brief.
190
+ - Write each brief per `.agents/skills/orkestrel-dispatch/SKILL.md` § Write the brief. A brief file lives at `tmp/units/<unit>-brief.md` (native) or `tmp/<bench>/<unit>-brief.md` (bench), and a re-run at `<unit>-brief-<n>.md`.
162
191
  - State in every brief: role and engine, one objective, the evidence slice, owned and off-limits files, the acceptance criteria cheap-first, the return shape, the deviation contract, and that the executor performs the work itself and spawns nothing.
163
192
  - Capture a campaign unit's report to `tmp/units/<unit>-report.md`, or read the bench's last-message file.
164
193
  - Launch every multi-minute command as a harness-tracked background command through `node .agents/skills/orkestrel-dispatch/scripts/launch.ts` under a cap sized from prior runs, and read a bench answer with that skill's `scripts/result.ts`. Every launcher, probe, and instrument is a TypeScript file run by `node`, never a shell or Python script. Read liveness from the artifact the work produces. Kill only a recorded process id, never a pattern, and confirm the process and its descendants are gone before another writer takes its files.
@@ -182,24 +211,26 @@ The Orchestrator triages: obvious correction → tighten and re-dispatch; missin
182
211
 
183
212
  ## Cleanup
184
213
 
185
- - When a unit is accepted, carry its measurements, unresolved findings, and any instrument worth keeping into the commit message, a tracked design, or a test, then delete its `tmp/units/<unit>-*` files and its bench journals.
214
+ - When a unit is accepted, carry its measurements, unresolved findings, and any instrument worth keeping into the commit message, a tracked design, or a test, then delete its `tmp/units/<unit>-*` files and its bench journals. Before a unit's files go, carry its effective brief (objective, corrections, source identity, rulings) into the tracked plan or the acceptance commit. Delete or move no path a live lane's brief names.
186
215
  - `.claude/rules/campaign.md` owns what `.orkestrel/` holds and when each file goes; git history is the archive.
187
216
  - Delete a probe from the source tree before the unit returns; promote a probe that settled a claim into a test.
188
217
  - Run `node .agents/skills/orkestrel-dispatch/scripts/sweep.ts --report` at acceptance; it lists leftovers under `tmp/` and `.orkestrel/` by age and deletes nothing. Run the same script with `--tmp` after the last live unit exits; it deletes files older than its threshold and refuses while a journal is changing.
189
- - A campaign artifact never lives in the package it is about.
218
+ - A campaign artifact lives under `.orkestrel/<package>/`, never under the package's source, test, or distribution tree.
190
219
 
191
220
  ## Cache
192
221
 
193
222
  - Keep the model, options, tools, and instruction prefix unchanged while they fit the task. Check the active provider and harness before claiming that a change preserves or invalidates caching. Treat a continued session as retained context, not as a cache hit; a cache hit depends on the provider, the prefix, and the retention window. Claim a saving only from recorded usage.
194
223
  - Keep the always-on instruction set small and load supporting rules by scope. When instructions change, use the harness's documented reload mechanism and tell affected agents what changed. Do not assume a file edit updates an active session.
195
- - Give reading and driver roles no instruction files (`omitClaudeMd` in Claude Code) when their charter carries the floor they need.
224
+ - Give reading and driver roles no instruction files when their charter carries the floor they need.
196
225
  - Route bulk reads per § Routing and keep only the distillate.
197
226
 
198
227
  ## Acceptance
199
228
 
200
229
  - No writer's and no bench's self-assessment is authoritative. The Orchestrator accepts after the review and the gates the size gate names.
230
+ - An accepted claim or gate row names the independent receipt (command, input identity, exit) that closed it; a writer's report states implementation status and never closure.
201
231
  - Accept when the exit criterion is met and the gates are green. Reopening an accepted criterion is the user's instruction, not a reviewer's finding.
232
+ - Before calling work ready, try the shipped result as the user will, on the build they run, and report what did not run.
202
233
  - Substitute an engine only when the same session recorded the bench dark; name the fallback in the brief and the carrying commit.
203
234
  - In a repository whose package the registry already serves, commit on the main branch and push it only on the user's go-ahead.
204
- - Commit by explicit paths. Before a push, read `git log @{upstream}..` and push only commits meant to publish together; hold a push while a commit depends on an unpublished local pin.
235
+ - Commit by explicit paths. Before a push, read `git log @{upstream}..` and push only commits meant to publish together; hold a push while a commit depends on an unpublished local pin. Push a commit only after the gate run on that commit, or record the push as ungated.
205
236
  - Publishing is the user's decision and credential; `orkestrel-publish` owns the release procedure, the bump rules, and the dependency-tarball rules.
@@ -9,12 +9,16 @@ The commands an agent runs on a Windows host, and their quirks. `.claude/rules/p
9
9
  - Kill a recorded tree with `taskkill /pid <pid> /T /F`. `launch.ts` kills the tree at its cap on this host.
10
10
  - Detach a run from the session's process tree with WMI: `Invoke-CimMethod -ClassName Win32_Process -MethodName Create -Arguments @{ CommandLine = '<command>' }`. The run starts under `WmiPrvSE.exe`, outside the session's job and process tree (verified 2026-10-16). It receives the user's registry environment, not the caller's, so pass each value it needs on its command line. Record its journal and pid-file paths before the launch.
11
11
  - Spawn npm without a shell through its JavaScript entry: `node <node directory>/node_modules/npm/bin/npm-cli.js <arguments>`. A `.cmd` shim cannot start without a shell, and `launch.ts` reports exit `127`.
12
+ - `cmd /c` keeps a trailing space inside `set VAR=value && …`; write `set "VAR=value"`.
13
+ - Deliver a console Ctrl+C through `WScript.Shell` `AppActivate` then `SendKeys('^c')` after confirming activation when no computer-use tool is available.
14
+ - A `rename` onto a freshly hard-linked file can return `EPERM` while a scanner holds it.
12
15
  - A child CLI can leak orphaned `conhost.exe` processes and open console windows. Count them with `Get-Process conhost` at each launch while such a lane is live.
13
16
 
14
17
  ## PowerShell 5.1
15
18
 
16
19
  - Chain with `;` and gate with `if ($?) { ... }`; `&&` and `||` are parse errors.
17
20
  - Edit text with the harness's file tools, never `Get-Content` piped through `-replace` into `Set-Content`. `-replace` matches case-insensitively, `Set-Content` writes the ANSI code page, `Set-Content -Encoding utf8` and `Out-File` write a byte-order mark, and the round trip turns non-ASCII punctuation into mojibake.
21
+ - Write a JSON file a tool must parse to a new path through the file tool or `[IO.File]::WriteAllText` with a BOM-free `UTF8Encoding`.
18
22
  - Write a commit message to a file with the harness's file tool and commit with `git commit -F <file>`. A here-string carrying double quotes splits into stray arguments, and a byte-order mark in the file lands in the commit subject.
19
23
  - Never pipe a command whose exit code gates the next step, a `node` probe included, into `Select-Object -First` or `-Last`. The cut pipeline kills the command early and leaves a misleading `$LASTEXITCODE`. Run it bare, read `$LASTEXITCODE`, or decide from `git status --porcelain` and `git rev-parse`.
20
24
  - PowerShell drops a bare `--` before a native command, so `npm run <script> -- <arguments>` loses the separator. Run the tool's own entry with an argument array, or call `npm.cmd`.
@@ -26,12 +30,13 @@ The commands an agent runs on a Windows host, and their quirks. `.claude/rules/p
26
30
  - `Get-FileHash` prints uppercase hexadecimal, and Node's `createHash` prints lowercase. Compare digests case-insensitively.
27
31
  - Redirecting a native command's standard error with `2>&1` wraps each line in an error record and sets `$?` to false on success. Let the harness capture standard error.
28
32
 
29
- ## Harness approval on this host
33
+ ## Release packs and terminals
30
34
 
31
- - Keep every program-carrying command in a TypeScript file run by `node`. A heredoc, `node -e`, an `&&` chain, or a `${...}` argument trips the harness's approval classifier; a one-shot read-only check can run inline.
32
- - Run a `Remove-Item` as its own short command with a relative path. The classifier blocks a command that pairs it with a path-like literal elsewhere, such as a `'\'` string. Verify with `git status`.
35
+ - Git Bash ships no `script` binary, so an agent cannot give npm a TTY there; the operator runs a browser-authorized `npm login` or `npm publish` in a real terminal.
36
+ - A pack made on Windows records `0644` for an entry git tracks as `100755`. Compare release packs between hosts by paths, sizes, and content hashes, and report executable-mode differences separately.
33
37
 
34
38
  ## Deleting untracked files
35
39
 
36
40
  - When a sweep you run by hand deletes untracked files, send them to the Recycle Bin through a shell file operation with undo, never permanent deletion, and report what went there. Leave emptying the bin to the user; hand them `Clear-RecycleBin -Force`.
41
+ - `DeleteDirectory(..., SendToRecycleBin)` fails on an npm-installed tree; recycle each file with `DeleteFile` and remove the emptied folders.
37
42
  - Remove a worktree with `git worktree remove`. Delete a leftover `node_modules` junction without recursing into its target.
@@ -19,17 +19,15 @@ Invoke `$orkestrel-harden` for each package implementation unit. This skill owns
19
19
 
20
20
  ## Execute the campaign
21
21
 
22
- 1. **Create one registry.** Record the campaign goal, authoritative session, repositories, branches, dirty state, declared/resolved versions, dependency edges, guide state, write scope, and exclusions.
23
- 2. **Bound context.** Scout only packages and consumers relevant to the behavior. Use registry/canonical sources for the rest of the fleet.
24
- 3. **Trace the real flow.** Follow public types and one concrete consumer end to end. Identify where ergonomics break and which layer owns the reusable mechanism.
25
- 4. **Assign ownership.** Keep core host-independent. Keep server free of browser assumptions and browser free of Node assumptions. Keep product policy in applications.
26
- 5. **Judge downstream friction.** Treat consumer gymnastics as evidence, not automatic proof. Decide whether the consumer misused a coherent API or exposed a general upstream defect.
27
- 6. **Design in dependency order.** Update guides and types first in the lowest owning package, then update every affected consumer atomically. Do not add compatibility shims.
28
- 7. **Harden each unit.** Apply `$orkestrel-harden` to every touched package with the exact owned scope and acceptance criteria.
29
- 8. **Prove the round trip.** Put integration coverage at the highest package that can exercise the real composed behavior. Retain focused deterministic coverage in lower packages.
30
- 9. **Validate local changes safely.** When necessary, use built artifacts in an isolated consumer and prove the resolved graph. Restore temporary manifest/lockfile state.
31
- 10. **Verify topologically.** Run local gates for touched packages from dependencies to dependents and inspect relevant generated outputs.
32
- 11. **Review the campaign.** Run the review `.agents/orchestration.md` § Size gate names on the integrated result, with one cross-package architecture and ergonomics pass over the whole graph. Do not review each unit separately.
22
+ Run `.agents/orchestration.md` § Campaign for the phases, the review, and the gates, and `$orkestrel-harden` for each package unit. Record the live state of every relevant repository per [fleet.md](references/fleet.md) § Verify live state. This skill adds:
23
+
24
+ 1. **Trace the real flow.** Follow public types and one concrete consumer end to end. Identify where ergonomics break and which layer owns the reusable mechanism.
25
+ 2. **Assign ownership.** Keep core host-independent. Keep server free of browser assumptions and browser free of Node assumptions. Keep product policy in applications.
26
+ 3. **Judge downstream friction.** Treat consumer gymnastics as evidence, not automatic proof. Decide whether the consumer misused a coherent API or exposed a general upstream defect.
27
+ 4. **Prove the round trip.** Put integration coverage at the highest package that can exercise the real composed behavior. Retain focused deterministic coverage in lower packages.
28
+ 5. **Validate local changes safely.** When necessary, use built artifacts in an isolated consumer and prove the resolved graph, per [fleet.md](references/fleet.md) § Validate local packages.
29
+
30
+ The review is one cross-package architecture and ergonomics pass over the whole graph, not one per unit.
33
31
 
34
32
  ## Accept the result
35
33
 
@@ -40,7 +38,6 @@ Completion requires:
40
38
  - every changed public contract, consumer, test, and guide is aligned;
41
39
  - real round-trip behavior is proven at the correct layer;
42
40
  - package versions/ranges/guides are reported from live evidence;
43
- - every touched repository's requested local gates are green;
44
- - no in-scope work is deferred.
41
+ - every touched repository's requested local gates are green.
45
42
 
46
43
  Report the final package graph, ownership decisions, shipped behavior, exact verification, residual risk, and any user decision still required.
@@ -1,5 +1,7 @@
1
1
  # Fleet alignment
2
2
 
3
+ This reference governs fleet alignment across repositories: verifying live state, planning dependency order, and verifying consumers.
4
+
3
5
  ## Verify live state
4
6
 
5
7
  For every relevant repository, record:
@@ -29,19 +31,11 @@ Keep one authoritative session per package. Re-establish live state before actin
29
31
 
30
32
  ## Validate local packages
31
33
 
32
- When a dependent must verify a local package:
33
-
34
- - build the dependency;
35
- - consume its built artifact only in an isolated or explicitly authorized consumer;
36
- - prove every relevant resolution points to the intended artifact;
37
- - run the dependent's targeted then final local gates;
38
- - restore temporary overrides, manifest edits, and lockfile state.
39
-
40
- Do not treat an undeclared source-checkout import as consumer proof.
34
+ When a dependent must verify a local package, pin it to a built tarball in an isolated or explicitly authorized consumer, run the dependent's targeted then final local gates, and restore the pin, per `.agents/skills/orkestrel-publish/references/release.md` § Fix a dependency before it publishes. Do not treat an undeclared source-checkout import as consumer proof.
41
35
 
42
36
  ## Align guides and verification
43
37
 
44
- Refresh dependency guides through the current scaffold workflow when the owning repository and scope permit it. Report drift rather than hand-editing a vendored guide whose canonical source lives elsewhere.
38
+ Refresh dependency guides with `scaffold catalog` when the owning repository and scope permit it. Report drift rather than hand-editing a vendored guide whose canonical source lives elsewhere.
45
39
 
46
40
  Run package-local verification from dependencies to dependents. A passing lower package does not prove its consumer; a passing consumer does not replace lower-package invariants.
47
41
 
@@ -42,6 +42,6 @@ Place each proof at the highest useful layer:
42
42
 
43
43
  Use the actual packages and transports. Use temporary resources or protocol-faithful fixture servers for deterministic network boundaries. Use the real external service when its behavior is the claim. Never simulate an owned package with a mock or fake.
44
44
 
45
- When the claim is that a foreign client can consume the stack — an editor, an agent CLI, a third-party protocol client — one representative real client of that class drives the surface end to end before the claim ships. Record the exact commands, the authentication and approval model that client needed, and every part of the surface it could not reach. Protocol-level tests prove the protocol; only the client proves the integration.
45
+ Prove a claim that a foreign client can consume the stack per `.agents/skills/orkestrel-harden/references/hardening.md` § Use real implementations.
46
46
 
47
47
  Test both successful composition and contract disagreement: invalid options, unavailable capability, partial failure, abort/cleanup, version mismatch, and lifecycle ordering where applicable.