@unerr-ai/unerr 0.0.0-beta.4 → 0.0.0-beta.5

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/README.md CHANGED
@@ -166,6 +166,7 @@ Works with any MCP-compatible agent. 15 supported out of the box.
166
166
  - **Convention awareness** — auto-detected naming, structure, and import patterns injected into agent context.
167
167
  - **Semantic search** — `semantic_search` finds conceptually similar code even with different naming. `find_similar` discovers related entities by embedding distance.
168
168
  - **Business context** — `get_business_context` explains why code exists — purpose, feature area, taxonomy.
169
+ - **Tool adoption nudging** — 5-layer reinforcement system ensures agents consistently use unerr MCP tools over built-in alternatives. Exec nudges, hook interception, instruction-level enforcement, and skill reminders — all automatic, zero config.
169
170
 
170
171
  ### Compounding (session 2+)
171
172
 
@@ -249,7 +250,9 @@ AI Agent (Claude Code / Cursor / VS Code / Windsurf / any MCP client)
249
250
  │ ├── Token budgeter (prevents context rot)
250
251
  │ └── Compression engine (11 strategies, 645+ classifiers)
251
252
 
252
- └── CLI hooks ──→ Output compression ──→ Graph-aware prioritization
253
+ ├── CLI hooks ──→ Output compression ──→ Graph-aware prioritization
254
+
255
+ └── Tool nudging ──→ 5-layer adoption reinforcement (hooks, exec, instructions)
253
256
  ```
254
257
 
255
258
  One process. Fully local. No API keys. No network calls. No cloud.