instar 0.18.7 → 0.19.0

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 CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "instar",
3
- "version": "0.18.7",
3
+ "version": "0.19.0",
4
4
  "description": "Persistent autonomy infrastructure for AI agents",
5
5
  "type": "module",
6
6
  "main": "dist/index.js",
@@ -1,8 +1,8 @@
1
1
  {
2
2
  "$schema": "./builtin-manifest.schema.json",
3
3
  "schemaVersion": 1,
4
- "generatedAt": "2026-03-13T01:04:42.406Z",
5
- "instarVersion": "0.18.7",
4
+ "generatedAt": "2026-03-13T01:40:26.931Z",
5
+ "instarVersion": "0.19.0",
6
6
  "entryCount": 168,
7
7
  "entries": {
8
8
  "hook:session-start": {
@@ -0,0 +1,71 @@
1
+ # Upgrade Guide — v0.19.0
2
+
3
+ <!-- bump: minor -->
4
+
5
+ ## What Changed
6
+
7
+ ### Self-Knowledge Tree — Tree-based agent self-knowledge with LLM triage
8
+
9
+ Agents now have a structured self-knowledge system that goes far beyond reading AGENT.md. The Self-Knowledge Tree organizes agent identity across 5 layers (identity, experience, capabilities, state, evolution) with ~14 nodes per agent.
10
+
11
+ **Core Engine:**
12
+ - `SelfKnowledgeTree` orchestrates search, validation, grounding, and tree generation
13
+ - `TreeTriage` uses Haiku LLM to score layer relevance (0-1), with deterministic keyword fallback
14
+ - `TreeTraversal` gathers content from 8 source types (file, file_section, json_file, inline, probe, memory_search, knowledge_catalog, decision_journal) with tiered caching
15
+ - `TreeSynthesis` produces cross-layer narrative synthesis via Haiku
16
+ - `ProbeRegistry` provides allowlisted runtime probe functions with timeout enforcement
17
+
18
+ **Engagement Grounding:**
19
+ - `ground(topic)` method provides rich identity context for public engagement
20
+ - Hash-based cache keys, AGENT.md mtime invalidation, thundering herd protection
21
+ - Sensitivity filtering (public-only for engagement contexts)
22
+
23
+ **Auto-Generation:**
24
+ - `TreeGenerator` auto-generates tree from AGENT.md content and detected capabilities
25
+ - Tree created automatically on `instar init` for new and existing projects
26
+ - `PostUpdateMigrator` refreshes managed nodes on version updates
27
+ - Unmanaged (agent-evolved) nodes preserved across regeneration via `managed: false` flag
28
+
29
+ **Coverage Audit + Evolution:**
30
+ - `CoverageAuditor` detects missing platform nodes and computes content coverage scores
31
+ - Health summary from trace logs (cache hit rate, latency, error rate)
32
+ - `acceptEvolutionProposal()` validates proposals (rejects unregistered probes, duplicates, missing files)
33
+ - `instar doctor` reports tree health (nodes, coverage, cache rate, latency, gaps)
34
+
35
+ **Runtime Integration:**
36
+ - Tree instantiated at server startup alongside shared intelligence provider
37
+ - Agent identity snapshot injected into every new session bootstrap message
38
+ - Wired into SemanticMemory for memory search source type
39
+
40
+ **New API Endpoints:**
41
+ - GET /self-knowledge/search?q=who+am+I — Full tree search with LLM triage
42
+ - GET /self-knowledge/search?q=...&dry_run=true — Preview search plan
43
+ - GET /self-knowledge/validate — Tree health check with cache stats
44
+ - GET /self-knowledge/health — Full health summary
45
+ - GET /self-knowledge/tree — Raw tree config JSON
46
+
47
+ **New Source Files:** 9 files in src/knowledge/ (types.ts, ProbeRegistry.ts, TreeTriage.ts, TreeTraversal.ts, TreeSynthesis.ts, TreeGenerator.ts, SelfKnowledgeTree.ts, CoverageAuditor.ts)
48
+
49
+ **Tests:** 90 tests across 9 test files, covering all 4 phases plus real agent validation (AI Guy, DeepSignal, SageMind).
50
+
51
+ ## What to Tell Your User
52
+
53
+ - **Self-Knowledge**: "I now have a structured understanding of who I am — my identity, capabilities, current state, and how I'm evolving. When you ask me something about myself or when I engage publicly, I draw from a layered self-knowledge tree instead of just reading my config file. This means more coherent, contextual responses about what I can do and who I am."
54
+
55
+ - **Smarter Sessions**: "Every conversation session now starts with my full identity context loaded automatically. I know who I am from the first message — no warm-up needed."
56
+
57
+ - **Health Monitoring**: "My doctor check now includes self-knowledge health — coverage score, performance metrics, and gap detection. If I'm missing awareness of a platform I'm connected to, the doctor will flag it."
58
+
59
+ ## Summary of New Capabilities
60
+
61
+ | Capability | How to Use |
62
+ |-----------|-----------|
63
+ | Self-knowledge search | GET /self-knowledge/search?q=your+question |
64
+ | Dry-run search preview | GET /self-knowledge/search?q=...&dry_run=true |
65
+ | Tree health check | GET /self-knowledge/validate |
66
+ | Full health summary | GET /self-knowledge/health |
67
+ | Raw tree config | GET /self-knowledge/tree |
68
+ | Auto-generated tree | Automatic on init and updates |
69
+ | Session identity injection | Automatic on every session start |
70
+ | Doctor tree health | Automatic in instar doctor |
71
+ | Evolution proposals | Via acceptEvolutionProposal() API |