@tangle-network/agent-knowledge 14.0.1 → 14.0.3

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/CHANGELOG.md CHANGED
@@ -1,5 +1,17 @@
1
1
  # Changelog
2
2
 
3
+ ## 14.0.3 — 2026-09-07
4
+
5
+ The Eval peer range admits both 0.174 and 0.175, restoring compatibility with Eval 0.174 consumers.
6
+ Runtime integration examples use the maintained package root import.
7
+
8
+ ## 14.0.2 — 2026-09-07
9
+
10
+ ### Changed
11
+
12
+ - Requires Eval `>=0.175.0 <0.176.0`.
13
+ Eval 0.175.0 reads Runtime's `result.json` and `failure.json` as a run's terminal record; Knowledge's own behavior is unchanged.
14
+
3
15
  ## 14.0.1 — 2026-09-07
4
16
 
5
17
  ### Fixed
package/README.md CHANGED
@@ -4,7 +4,7 @@ Build, search, test, and improve source-backed knowledge bases in TypeScript.
4
4
 
5
5
  The package manages source records, Markdown knowledge, indexes, retrieval tests, candidate workspaces, and memory adapters.
6
6
  It does not browse, call models, or run agents.
7
- Supply application callbacks for those decisions, or use `@tangle-network/agent-runtime/knowledge` to run them with agents.
7
+ Supply application callbacks for those decisions, or use `@tangle-network/agent-runtime` to run them with agents.
8
8
 
9
9
  ## Install
10
10
 
@@ -29,7 +29,7 @@ Requires Node.js 20.19 or later.
29
29
  | Compare providers or optimize memory configuration | `AgentMemoryAdapter`, `runAgentMemoryImprovement` | `/memory` |
30
30
  | Read from external authorities | `KnowledgeSource` and source adapters | `/sources` |
31
31
  | Run retrieval, answer, KB, or memory benchmark cases | `runKnowledgeBenchmarkSuite` | `/benchmarks` |
32
- | Use live research or coding agents | `runKnowledgeImprovementJob` | `@tangle-network/agent-runtime/knowledge` |
32
+ | Use live research or coding agents | `runKnowledgeImprovementJob` | `@tangle-network/agent-runtime` |
33
33
 
34
34
  ## Create and search a knowledge base
35
35
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@tangle-network/agent-knowledge",
3
- "version": "14.0.1",
3
+ "version": "14.0.3",
4
4
  "description": "Build, search, evaluate, and improve source-backed knowledge bases.",
5
5
  "homepage": "https://github.com/tangle-network/agent-knowledge#readme",
6
6
  "repository": {
@@ -84,14 +84,14 @@
84
84
  "zod": "4.5.4"
85
85
  },
86
86
  "peerDependencies": {
87
- "@tangle-network/agent-eval": ">=0.174.0 <0.175.0",
87
+ "@tangle-network/agent-eval": ">=0.174.0 <0.176.0",
88
88
  "@tangle-network/agent-interface": "^2.0.0"
89
89
  },
90
90
  "devDependencies": {
91
91
  "@arethetypeswrong/cli": "^0.18.5",
92
92
  "@biomejs/biome": "^2.5.11",
93
93
  "@neo4j-labs/agent-memory": "0.4.1",
94
- "@tangle-network/agent-eval": "0.174.0",
94
+ "@tangle-network/agent-eval": "0.175.0",
95
95
  "@tangle-network/agent-interface": "2.0.0",
96
96
  "@types/node": "^26.4.0",
97
97
  "mem0ai": "3.1.7",
@@ -46,7 +46,7 @@ The agent's current answer is not a gold answer.
46
46
  It does not own model choice, prompts, browsing, agent scheduling, product authorization, or product storage transactions.
47
47
 
48
48
  Supply callbacks for research, retrieval, answer generation, and scoring.
49
- Use `@tangle-network/agent-runtime/knowledge` when those callbacks should run agents.
49
+ Use `@tangle-network/agent-runtime` when those callbacks should run agents.
50
50
  Use existing vector, graph, search, and memory systems through adapters instead of rebuilding their databases here.
51
51
 
52
52
  ## Build The Smallest Complete Path