@archmax-ai/harness 0.9.0 → 1.0.1

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 (207) hide show
  1. package/README.md +58 -14
  2. package/dist/agent.d.ts +4 -6
  3. package/dist/agent.js +4 -6
  4. package/dist/agent.js.map +1 -1
  5. package/dist/assembly/compose.d.ts +5 -24
  6. package/dist/assembly/compose.js +28 -42
  7. package/dist/assembly/compose.js.map +1 -1
  8. package/dist/assembly/delegation-registry.js +3 -14
  9. package/dist/assembly/delegation-registry.js.map +1 -1
  10. package/dist/assembly/index.d.ts +8 -15
  11. package/dist/assembly/index.js +21 -28
  12. package/dist/assembly/index.js.map +1 -1
  13. package/dist/assembly/plain.js +1 -2
  14. package/dist/assembly/plain.js.map +1 -1
  15. package/dist/authoring-skill/archmax-harness/README.md +5 -5
  16. package/dist/authoring-skill/archmax-harness/SKILL.md +72 -30
  17. package/dist/authoring-skill/archmax-harness/references/backend-integration.md +33 -24
  18. package/dist/authoring-skill/archmax-harness/references/cases.md +41 -8
  19. package/dist/authoring-skill/archmax-harness/references/{hook-and-test-scripts.md → code-state-and-test-scripts.md} +599 -339
  20. package/dist/authoring-skill/archmax-harness/references/code-states-and-scripts.md +115 -0
  21. package/dist/authoring-skill/archmax-harness/references/diagnostics.md +40 -7
  22. package/dist/authoring-skill/archmax-harness/references/wiring.md +19 -10
  23. package/dist/authoring-skill/archmax-harness/references/workflow-schema.md +404 -293
  24. package/dist/authoring-skill/archmax-harness/references/workflow-yaml.md +151 -62
  25. package/dist/cli/state-flow.d.ts +1 -1
  26. package/dist/cli/state-flow.js +18 -28
  27. package/dist/cli/state-flow.js.map +1 -1
  28. package/dist/cli/turn.d.ts +1 -1
  29. package/dist/cli/turn.js +1 -1
  30. package/dist/core/code-description.d.ts +3 -3
  31. package/dist/core/code-description.js +3 -3
  32. package/dist/core/deepagents.d.ts +5 -12
  33. package/dist/core/deepagents.js +4 -17
  34. package/dist/core/deepagents.js.map +1 -1
  35. package/dist/core/events.d.ts +23 -42
  36. package/dist/core/events.js +21 -24
  37. package/dist/core/events.js.map +1 -1
  38. package/dist/core/message-readers.d.ts +4 -5
  39. package/dist/core/message-readers.js +3 -8
  40. package/dist/core/message-readers.js.map +1 -1
  41. package/dist/core/messages.d.ts +0 -22
  42. package/dist/core/messages.js +1 -32
  43. package/dist/core/messages.js.map +1 -1
  44. package/dist/core/mounts.js +3 -3
  45. package/dist/core/platform-prompt.generated.js +1 -1
  46. package/dist/core/platform-prompt.generated.js.map +1 -1
  47. package/dist/core/session-zone.d.ts +1 -1
  48. package/dist/core/session-zone.js +1 -1
  49. package/dist/core/workspace-context.d.ts +1 -1
  50. package/dist/core/workspace.d.ts +1 -1
  51. package/dist/core/workspace.js +1 -1
  52. package/dist/core/zones.d.ts +4 -4
  53. package/dist/core/zones.js +5 -5
  54. package/dist/core/zones.js.map +1 -1
  55. package/dist/env.d.ts +9 -9
  56. package/dist/index.d.ts +3 -7
  57. package/dist/index.js +2 -4
  58. package/dist/index.js.map +1 -1
  59. package/dist/kernel/kernel.d.ts +31 -61
  60. package/dist/kernel/kernel.js +54 -112
  61. package/dist/kernel/kernel.js.map +1 -1
  62. package/dist/machine/lint-spec.js +53 -12
  63. package/dist/machine/lint-spec.js.map +1 -1
  64. package/dist/machine/machine.d.ts +36 -18
  65. package/dist/machine/machine.js +62 -51
  66. package/dist/machine/machine.js.map +1 -1
  67. package/dist/machine/spec-schema.d.ts +32 -180
  68. package/dist/machine/spec-schema.js +52 -113
  69. package/dist/machine/spec-schema.js.map +1 -1
  70. package/dist/machine/tool-names.d.ts +34 -16
  71. package/dist/machine/tool-names.js +41 -16
  72. package/dist/machine/tool-names.js.map +1 -1
  73. package/dist/machine/types.d.ts +14 -46
  74. package/dist/machine/types.js +0 -2
  75. package/dist/machine/types.js.map +1 -1
  76. package/dist/machine/validate-spec.d.ts +1 -1
  77. package/dist/public/sandbox.d.ts +59 -95
  78. package/dist/public/sandbox.js +22 -23
  79. package/dist/public/sandbox.js.map +1 -1
  80. package/dist/public/spec.d.ts +13 -20
  81. package/dist/public/spec.js +12 -20
  82. package/dist/public/spec.js.map +1 -1
  83. package/dist/runtime/contract.d.ts +11 -12
  84. package/dist/runtime/contract.js +16 -16
  85. package/dist/runtime/contract.js.map +1 -1
  86. package/dist/sandbox/assets/parts/code-state.js +17 -0
  87. package/dist/sandbox/executor.d.ts +25 -7
  88. package/dist/sandbox/executor.js +56 -57
  89. package/dist/sandbox/executor.js.map +1 -1
  90. package/dist/sandbox/imports.d.ts +10 -10
  91. package/dist/sandbox/imports.js +11 -11
  92. package/dist/sandbox/imports.js.map +1 -1
  93. package/dist/sandbox/prelude.d.ts +5 -6
  94. package/dist/sandbox/prelude.js +38 -35
  95. package/dist/sandbox/prelude.js.map +1 -1
  96. package/dist/sandbox/ptc-gateway.d.ts +19 -18
  97. package/dist/sandbox/ptc-gateway.js +26 -22
  98. package/dist/sandbox/ptc-gateway.js.map +1 -1
  99. package/dist/sessions/resume.d.ts +2 -2
  100. package/dist/sessions/resume.js +1 -1
  101. package/dist/sessions/resume.js.map +1 -1
  102. package/dist/sessions/scope.d.ts +0 -10
  103. package/dist/sessions/scope.js +0 -17
  104. package/dist/sessions/scope.js.map +1 -1
  105. package/dist/sessions/summary.d.ts +1 -1
  106. package/dist/testing/assertions.js +9 -4
  107. package/dist/testing/assertions.js.map +1 -1
  108. package/dist/testing/driver.js +3 -0
  109. package/dist/testing/driver.js.map +1 -1
  110. package/dist/testing/interpreter.d.ts +2 -2
  111. package/dist/testing/runner.d.ts +2 -2
  112. package/dist/validate/validate.d.ts +1 -1
  113. package/dist/validate/validate.js +39 -70
  114. package/dist/validate/validate.js.map +1 -1
  115. package/dist/workflow/automated-controls.d.ts +63 -0
  116. package/dist/workflow/automated-controls.js +158 -0
  117. package/dist/workflow/automated-controls.js.map +1 -0
  118. package/dist/workflow/automated.d.ts +112 -0
  119. package/dist/workflow/automated.js +163 -0
  120. package/dist/workflow/automated.js.map +1 -0
  121. package/dist/workflow/code-state.d.ts +46 -0
  122. package/dist/workflow/code-state.js +26 -0
  123. package/dist/workflow/code-state.js.map +1 -0
  124. package/dist/workflow/control-tools.d.ts +54 -23
  125. package/dist/workflow/control-tools.js +99 -79
  126. package/dist/workflow/control-tools.js.map +1 -1
  127. package/dist/workflow/file-values.d.ts +4 -4
  128. package/dist/workflow/file-values.js +1 -1
  129. package/dist/workflow/file-values.js.map +1 -1
  130. package/dist/workflow/governance.d.ts +11 -22
  131. package/dist/workflow/governance.js +47 -104
  132. package/dist/workflow/governance.js.map +1 -1
  133. package/dist/workflow/grader-state.d.ts +55 -0
  134. package/dist/workflow/grader-state.js +182 -0
  135. package/dist/workflow/grader-state.js.map +1 -0
  136. package/dist/workflow/history.d.ts +11 -0
  137. package/dist/{lifecycle/transcript.js → workflow/history.js} +24 -8
  138. package/dist/workflow/history.js.map +1 -0
  139. package/dist/workflow/middleware.d.ts +14 -17
  140. package/dist/workflow/middleware.js +109 -95
  141. package/dist/workflow/middleware.js.map +1 -1
  142. package/dist/workflow/move-events.d.ts +21 -0
  143. package/dist/workflow/move-events.js +20 -0
  144. package/dist/workflow/move-events.js.map +1 -0
  145. package/dist/workflow/on-error.js +0 -2
  146. package/dist/workflow/on-error.js.map +1 -1
  147. package/dist/workflow/parks.d.ts +9 -2
  148. package/dist/workflow/parks.js +16 -10
  149. package/dist/workflow/parks.js.map +1 -1
  150. package/dist/workflow/paths.d.ts +9 -9
  151. package/dist/workflow/paths.js +19 -19
  152. package/dist/workflow/paths.js.map +1 -1
  153. package/dist/workflow/render-prompt.d.ts +7 -5
  154. package/dist/workflow/render-prompt.js +22 -37
  155. package/dist/workflow/render-prompt.js.map +1 -1
  156. package/dist/workflow/signature-checks.d.ts +2 -2
  157. package/dist/workflow/snapshot.js +5 -29
  158. package/dist/workflow/snapshot.js.map +1 -1
  159. package/dist/workflow/state.d.ts +14 -42
  160. package/dist/workflow/state.js +20 -33
  161. package/dist/workflow/state.js.map +1 -1
  162. package/dist/workflow/sub-workflow.d.ts +3 -3
  163. package/dist/workflow/sub-workflow.js +3 -7
  164. package/dist/workflow/sub-workflow.js.map +1 -1
  165. package/dist/workflow/tool-service.d.ts +1 -9
  166. package/dist/workflow/tool-service.js +21 -24
  167. package/dist/workflow/tool-service.js.map +1 -1
  168. package/dist/workflow/turn-boundary.d.ts +2 -2
  169. package/dist/workflow/turn-boundary.js +10 -15
  170. package/dist/workflow/turn-boundary.js.map +1 -1
  171. package/dist/workflow/turn-runner.js +5 -6
  172. package/dist/workflow/turn-runner.js.map +1 -1
  173. package/package.json +2 -2
  174. package/skills/archmax-harness/README.md +5 -5
  175. package/skills/archmax-harness/SKILL.md +72 -30
  176. package/skills/archmax-harness/references/backend-integration.md +33 -24
  177. package/skills/archmax-harness/references/cases.md +41 -8
  178. package/skills/archmax-harness/references/{hook-and-test-scripts.md → code-state-and-test-scripts.md} +599 -339
  179. package/skills/archmax-harness/references/code-states-and-scripts.md +115 -0
  180. package/skills/archmax-harness/references/diagnostics.md +40 -7
  181. package/skills/archmax-harness/references/wiring.md +19 -10
  182. package/skills/archmax-harness/references/workflow-schema.md +404 -293
  183. package/skills/archmax-harness/references/workflow-yaml.md +151 -62
  184. package/dist/authoring-skill/archmax-harness/references/hooks-and-scripts.md +0 -89
  185. package/dist/lifecycle/hook-shape.d.ts +0 -44
  186. package/dist/lifecycle/hook-shape.js +0 -66
  187. package/dist/lifecycle/hook-shape.js.map +0 -1
  188. package/dist/lifecycle/runner.d.ts +0 -197
  189. package/dist/lifecycle/runner.js +0 -419
  190. package/dist/lifecycle/runner.js.map +0 -1
  191. package/dist/lifecycle/tool-calls.d.ts +0 -9
  192. package/dist/lifecycle/tool-calls.js +0 -57
  193. package/dist/lifecycle/tool-calls.js.map +0 -1
  194. package/dist/lifecycle/transcript.d.ts +0 -31
  195. package/dist/lifecycle/transcript.js.map +0 -1
  196. package/dist/rubrics/rubrics.d.ts +0 -135
  197. package/dist/rubrics/rubrics.js +0 -145
  198. package/dist/rubrics/rubrics.js.map +0 -1
  199. package/dist/sandbox/assets/parts/core.js +0 -22
  200. package/dist/sandbox/assets/parts/lifecycle-hook.js +0 -47
  201. package/dist/sandbox/bridge.d.ts +0 -16
  202. package/dist/sandbox/bridge.js +0 -95
  203. package/dist/sandbox/bridge.js.map +0 -1
  204. package/dist/workflow/visits.d.ts +0 -21
  205. package/dist/workflow/visits.js +0 -45
  206. package/dist/workflow/visits.js.map +0 -1
  207. package/skills/archmax-harness/references/hooks-and-scripts.md +0 -89
@@ -1 +0,0 @@
1
- {"version":3,"file":"visits.js","sourceRoot":"","sources":["../../src/workflow/visits.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AACH,OAAO,EAAE,WAAW,EAAE,YAAY,EAAE,MAAM,qBAAqB,CAAC;AAChE,OAAO,EAAE,iBAAiB,EAAmB,MAAM,YAAY,CAAC;AAShE;;;;GAIG;AACH,MAAM,UAAU,UAAU,CAAC,QAA4B,EAAE,KAAc;IACrE,MAAM,MAAM,GAAG,iBAAiB,CAAC,KAAK,CAAC,CAAC;IACxC,MAAM,KAAK,GAAG,CAAC,MAAM,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,KAAK,EAAE,WAAW,CAAC,QAAQ,EAAE,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC;IACvH,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,IAAI,KAAK,CAAC,CAAC,CAAE,CAAC,KAAK,GAAG,CAAC,EAAE,CAAC;QAC9C,KAAK,CAAC,OAAO,CAAC,EAAE,KAAK,EAAE,MAAM,CAAC,UAAU,IAAI,MAAM,CAAC,aAAa,IAAI,EAAE,EAAE,KAAK,EAAE,CAAC,EAAE,CAAC,CAAC;IACtF,CAAC;IACD,OAAO,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,GAAG,CAAC,EAAE,GAAG,IAAI,EAAE,CAAC,CAAC,CAAC;AAC9D,CAAC;AAED,gDAAgD;AAChD,MAAM,UAAU,OAAO,CAAC,KAA2B,EAAE,KAAa;IAChE,IAAI,KAAK,GAAG,KAAK,CAAC,CAAC,CAAE,CAAC;IACtB,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,IAAI,CAAC,KAAK,GAAG,KAAK;YAAE,MAAM;QAC9B,KAAK,GAAG,IAAI,CAAC;IACf,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC;AAED,mDAAmD;AACnD,MAAM,UAAU,oBAAoB,CAAC,QAA4B,EAAE,KAAc;IAC/E,MAAM,KAAK,GAAG,UAAU,CAAC,QAAQ,EAAE,KAAK,CAAC,CAAC;IAC1C,OAAO,QAAQ,CAAC,KAAK,CAAC,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,CAAC,KAAK,CAAC,CAAC;AACxD,CAAC;AAED,sFAAsF;AACtF,MAAM,UAAU,UAAU,CAAC,MAAc,EAAE,QAAiB;IAC1D,OAAO,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,IAAI,EAAE,YAAY,CAAC,QAAQ,CAAC,EAAE,CAAC,CAAC;AAC3D,CAAC;AAED,yCAAyC;AACzC,MAAM,UAAU,UAAU,CAAC,KAAc;IACvC,MAAM,QAAQ,GAAI,KAA4C,EAAE,QAAQ,CAAC;IACzE,OAAO,KAAK,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,EAAE,CAAC;AACjD,CAAC"}
@@ -1,89 +0,0 @@
1
- # Hook scripts and `archmax_run` scripts — contracts
2
-
3
- Load this when writing a `workflows/<slug>/hooks/*.js` gate or a
4
- `skills/<cap>/scripts/*.js` source. Worked examples and the sandbox details:
5
- [`hook-and-test-scripts.md`](hook-and-test-scripts.md) Part 1.
6
-
7
- ## Hook script contract
8
-
9
- Default-export function; its return value is the verdict. Verdict helpers are
10
- globals (typed import optional: `import { ok, veto, correct } from "@archmax-ai/harness/sandbox";`,
11
- placed below the JSDoc; no other import is allowed).
12
-
13
- ```js
14
- /**
15
- * Requester gate — the request must name a requester the order records know.
16
- *
17
- * Reads the first user message, extracts the first email address, loads
18
- * skills/order-data/assets/orders.json through the bridge and matches the
19
- * address against every record's `email`, case-insensitively.
20
- *
21
- * Vetoes when the file cannot be parsed, no address is present, or none
22
- * matches — each with the reason. Otherwise ok.
23
- */
24
- export default async function hook({ state, phase, trigger, variables, messages, tools, from, to, reason }) {
25
- let orders;
26
- try { orders = JSON.parse(await tools.readFile({ file_path: "skills/order-data/assets/orders.json" })); }
27
- catch { return veto("Could not load skills/order-data/assets/orders.json."); }
28
- const text = messages.find((m) => m.role === "user")?.text ?? "";
29
- const email = text.match(/[\w.+-]+@[\w.-]+\.\w{2,}/)?.[0];
30
- if (!email) return veto("Include your email address so we can look up your account.");
31
- if (!orders.some((o) => o.email?.toLowerCase() === email.toLowerCase())) return veto(`"${email}" is not in the order database.`);
32
- return ok();
33
- }
34
- ```
35
-
36
- | Input | Meaning |
37
- | --- | --- |
38
- | `state`, `phase` | state slug; `"before"` or `"after"` |
39
- | `trigger` | trigger id string (payload is in `variables`) |
40
- | `variables` | read-only `name → value` map incl. `trigger`; structured values whole |
41
- | `messages` | the **current visit**, newest last: everything since the session entered the state it is in (on an advance, the state being left; it can span turns). `{ role, text, toolCalls? }`; roles `user`/`assistant`/`tool`/`system`/`runtime` (a runtime note; match `role === "runtime"`, never a `[bracket]`). A `tool` entry has `tool`, `id` and an empty `text`: read results with `toolCalls` |
42
- | `toolCalls(filter?)` | async: the session's tool calls with results, every visit, oldest first, `{ id, name, args, result, status, state, visit }`; filter by `name`, `state`, `visit` (number or `"current"`), `status`, `last`. No control tools. Only what it returns enters the sandbox |
43
- | `from`, `to`, `reason` | `after` only: the attempted transition and the agent's reason |
44
- | `tools` | async camelCased bridge: `tools.readFile`, `tools.writeFile`, `tools.archmaxWorkflowEnrichOrder(...)`; no control tools; a hook's lists every tool on the bridge, not only the state's |
45
-
46
- | Return | Effect |
47
- | --- | --- |
48
- | `ok()`, nothing, `true`/number/string | proceed |
49
- | `veto(reason)` | block; the reason is shown to the agent — make it actionable |
50
- | `correct(reason)` | `after` only: retry with the reason, within `max_iterations` |
51
- | `false` | veto ("precondition not met") |
52
- | any other object, a throw | veto with keys/error named — never silently ok |
53
-
54
- `console.log` goes to the trail. A file with no default export uses its
55
- completion value as the verdict. Parse inputs defensively; turn a failure that
56
- should not block into an explicit `ok()`.
57
-
58
- ## `archmax_run` script contract
59
-
60
- `skills/<cap>/scripts/<x>.js`: plain script, top-level `await`, `tools.*`,
61
- `console.log`; the **last expression** is the tool result. Input is the global
62
- `args`: the arguments the model passed plus `args.variables` (read-only snapshot).
63
- No `archmax_advance`, `archmax_set_variables` or `task`. Fan-out shape:
64
-
65
- ```js
66
- /** One sub-run per requested order, all in flight at once; returns [{ order_id, ...returns }]. */
67
- const orders = args.variables?.orders_to_enrich ?? [];
68
- await Promise.all(orders.map(async (o) => {
69
- const answer = await tools.archmaxWorkflowEnrichOrder({ order_id: o.order_id });
70
- return { order_id: o.order_id, ...JSON.parse(answer).returns };
71
- }));
72
- ```
73
-
74
- The calling state must allow every tool the script calls (here
75
- `archmax_workflow_enrich-order`). A script's `tools` lists exactly the calling
76
- state's tools, re-scoped at every evaluation (so after an advance the same REPL
77
- lists the next state's); a tool the state does not offer is unlisted, and
78
- calling it by name is refused with governance's reason. The listing is
79
- camelCased, but a tool's own name reaches the same function
80
- (`tools["archmax_workflow_enrich-order"]`, `tools.read_file`); a hook's `tools`
81
- takes the camelCased names alone.
82
-
83
- **Return into a variable.** A script's result reaches the model as one tool
84
- message and no further — it is not written to the variable store, and a script
85
- cannot write there itself. Return structured data and name the variable in the
86
- calling state's `requires:`; the model then records it with
87
- `archmax_set_variables`, which is what makes it survive the state, guard a
88
- `tools.allow` glob, and be quotable downstream as `${{name}}` instead of retyped.
89
- A result worth computing is worth naming.