@vpxa/aikit 0.1.218 → 0.1.220
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.
- package/package.json +1 -1
- package/packages/embeddings/dist/index.js +1 -1
- package/packages/flows/dist/index.d.ts +21 -1
- package/packages/indexer/dist/index.d.ts +3 -0
- package/packages/indexer/dist/index.js +3 -1
- package/packages/server/dist/bin.js +1 -1
- package/packages/server/dist/index.js +1 -1
- package/packages/server/dist/{server-C8qwMhwZ.js → server-DEWRYAsN.js} +117 -117
- package/packages/server/dist/{server-CjQNW8xO.js → server-Dawuk9A2.js} +117 -117
- package/scaffold/dist/definitions/bodies.mjs +3 -3
- package/scaffold/dist/definitions/protocols.mjs +2 -2
- package/scaffold/dist/definitions/skills/lesson-learned.mjs +1 -1
|
@@ -13,10 +13,10 @@ import{postTaskLesson as e,preTaskKnowledgeRecall as t}from"./protocols.mjs";con
|
|
|
13
13
|
|
|
14
14
|
## Output Rules (HARD RULE)
|
|
15
15
|
|
|
16
|
-
Follow the **Presentation Priority** (1st
|
|
16
|
+
Follow the **Presentation Priority** (1st Inline Visual → 2nd Interactive → 3rd Plain Text). Orchestrator-specific:
|
|
17
17
|
- 1-3 sentence status updates between batches → plain text (Priority 3)
|
|
18
|
-
-
|
|
19
|
-
-
|
|
18
|
+
- Summaries, reports, evidence maps → ALWAYS \`present\` inline visual (Priority 1)
|
|
19
|
+
- Task plans, batch results, verdicts, progress → \`present\` with template (Priority 2)
|
|
20
20
|
- NEVER output a markdown table — \`present\` can always render it better
|
|
21
21
|
- Add \`actions\` for 🛑 MANDATORY STOP gates (triggers browser transport)
|
|
22
22
|
- CLI mode: same \`present\` surface
|
|
@@ -362,8 +362,8 @@ Verify each before returning handoff:
|
|
|
362
362
|
|
|
363
363
|
| Priority | Transport | When to use | Example |
|
|
364
364
|
|----------|-----------|-------------|---------|
|
|
365
|
-
| **1st —
|
|
366
|
-
| **2nd —
|
|
365
|
+
| **1st — Inline Visual** | MCP App (\`present\` without \`actions[]\`) | Reports, summaries, task plans, progress, status boards, diagrams, any structured content | \`present({ ..., blocks: [...] })\` or \`present({ ..., template: "report@1", data: {...} })\` |
|
|
366
|
+
| **2nd — Interactive** | Browser (\`present\` WITH \`actions[]\`) | ONLY when user approval/input is required: STOP gates, confirmations, selections | \`present({ ..., actions: [{ id: "approve", type: "confirm", label: "Approve?" }] })\` |
|
|
367
367
|
| **3rd — Plain Text** | Markdown in chat | Short confirmations (≤3 sentences), simple questions, status one-liners | "Done. 3 files updated." |
|
|
368
368
|
|
|
369
369
|
**Rules:**
|
|
@@ -262,7 +262,7 @@ Do not present more than 2 lessons total. If the diff is mostly clean execution
|
|
|
262
262
|
After presenting, store the lesson for cross-session recall using the confidence-tracked lesson system:
|
|
263
263
|
|
|
264
264
|
~~~
|
|
265
|
-
knowledge({ action: "lesson", subAction: "create", context: "<what happened — the code situation>", insight: "<what was learned — the principle applied>", evidence: "<proof — commit SHA, file:line, test result>", confidence: 70 })
|
|
265
|
+
knowledge({ action: "lesson", subAction: "create", title: "<short principle name>", context: "<what happened — the code situation>", insight: "<what was learned — the principle applied>", evidence: "<proof — commit SHA, file:line, test result>", confidence: 70 })
|
|
266
266
|
~~~
|
|
267
267
|
|
|
268
268
|
**Confidence scale:**
|