@dzhechkov/harness-cli 0.3.28 → 0.3.29

package/README.md

@@ -365,6 +365,57 @@ dz dashboard    # Visual panel with all packages, adapters, skill packs
 
 ---
 
+## Self-Learning: JSONL vs AgentDB
+
+DZ Harness supports two memory backends for self-learning:
+
+```bash
+dz setup --target claude-code --preset devops                    # JSONL (default, lightweight)
+dz setup --target claude-code --preset devops --memory agentdb   # AgentDB (vector memory)
+```
+
+| Capability | JSONL (default) | AgentDB (`--memory agentdb`) |
+|------------|----------------|------------------------------|
+| **Session tracking** | Append-only JSONL log | HNSW vector store (.rvf) |
+| **Pattern storage** | `dz teach` → patterns.jsonl | `dz teach` → .rvf + agentdb_pattern_store |
+| **Search** | Keyword (grep) | **Semantic** (HNSW nearest-neighbor, cosine similarity) |
+| **Retrieval** | Sequential scan | **O(log n)** approximate nearest neighbor |
+| **Self-learning** | Frequency-based | **9 RL algorithms** + Thompson Sampling bandit |
+| **Memory tiers** | Flat file | **3-tier** (working → short-term → long-term) |
+| **Reflexion** | Reward scores (0-1) | Episodic memory (task + outcome + self-critique) |
+| **Causal reasoning** | No | **Cypher-like graph queries** (X caused Y) |
+| **Skill composition** | Manual (presets) | **Bandit-picked** skill chains (A→B→C) |
+| **Audit trail** | No | **Cryptographic attestation log** |
+| **Size** | ~0 KB | 4.6 MB (agentdb) |
+| **MCP tools** | 0 | **41 tools** (pattern, reflexion, causal, skill, hierarchy) |
+| **Dependencies** | None | agentdb (optional, via npx) |
+
+### AgentDB self-learning algorithms
+
+When using `--memory agentdb`, the following algorithms automatically tune search quality:
+
+1. **Thompson Sampling** — multi-armed bandit for ranking search results
+2. **UCB1** (Upper Confidence Bound) — exploration-exploitation balancing
+3. **EXP3** — adversarial bandit for non-stationary environments
+4. **Softmax** — temperature-based action selection
+5. **Epsilon-Greedy** — simple exploration with decay
+6. **Gradient Bandit** — preference-based action selection
+7. **Contextual Bandit** — context-aware ranking using features
+8. **REINFORCE** — policy gradient for complex reward landscapes
+9. **PPO-lite** — proximal policy optimization for stable learning
+
+The bandit automatically selects the best algorithm for your usage pattern — no manual tuning needed.
+
+### Three ways to use AgentDB
+
+| Command | What it does | Result |
+|---------|-------------|--------|
+| `dz setup --memory agentdb` | Creates .rvf, registers MCP server, configures hooks | Infrastructure ready |
+| `dz init --select agentdb-memory` | Installs SKILL.md teaching agent to use 41 MCP tools | Agent knows how |
+| Both (recommended) | Full setup: infrastructure + instructions | Agent can self-learn |
+
+---
+
 ## How it works
 
 - `dz init` compiles canonical skills from the [agentskills.io](https://agentskills.io) standard into the target platform's layout
@@ -541,4 +592,4 @@ npx @dzhechkov/p-replicator init
 
 ## Status
 
-`v0.3.27` — published on npm. Part of [DZ Harness Hub](https://github.com/djd1m/dz-harness-hub).
+`v0.3.29` — published on npm. Part of [DZ Harness Hub](https://github.com/djd1m/dz-harness-hub).

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@dzhechkov/harness-cli",
-  "version": "0.3.28",
+  "version": "0.3.29",
   "description": "The dz CLI — install AI skills for Claude Code, Codex, OpenCode, Hermes. 11 commands, 7 presets, 4 platform adapters.",
   "type": "module",
   "license": "MIT",
@@ -35,8 +35,8 @@
   ],
   "dependencies": {
     "@dzhechkov/harness-presets": "0.2.9",
-    "@dzhechkov/harness-core": "0.3.12",
-    "@dzhechkov/scout": "0.7.1"
+    "@dzhechkov/scout": "0.7.1",
+    "@dzhechkov/harness-core": "0.3.12"
   },
   "devDependencies": {
     "@types/node": "^25.6.0",