@nano-step/nano-brain 2026.5.266 → 2026.5.267
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 +39 -0
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -130,8 +130,46 @@ telemetry:
|
|
|
130
130
|
logging:
|
|
131
131
|
level: info
|
|
132
132
|
file: "" # empty = stdout only
|
|
133
|
+
|
|
134
|
+
summarization:
|
|
135
|
+
enabled: false # set to true to generate LLM summaries of harvested sessions
|
|
136
|
+
provider_url: "" # OpenAI-compatible endpoint, e.g. https://ai-proxy.example.com/v1
|
|
137
|
+
api_key: "" # or set NANO_BRAIN_SUMMARIZE_API_KEY env var
|
|
138
|
+
model: "nano-brain" # model name passed to the provider
|
|
139
|
+
max_tokens: 4096 # max tokens per LLM completion
|
|
140
|
+
concurrency: 3 # parallel map-phase LLM calls
|
|
141
|
+
output_dir: "~/.nano-brain/summaries" # directory for .md summary files
|
|
142
|
+
```
|
|
143
|
+
|
|
144
|
+
### Session Summarization
|
|
145
|
+
|
|
146
|
+
When `summarization.enabled: true`, nano-brain automatically generates structured markdown summaries of each harvested session using an OpenAI-compatible LLM provider. Summaries are:
|
|
147
|
+
|
|
148
|
+
- Written as `.md` files to `output_dir` (`{source}_{title-slug}_{YYYY-MM-DD}.md`)
|
|
149
|
+
- Stored in the vector DB under collection `session-summary` for semantic search
|
|
150
|
+
- Idempotent — unchanged sessions are skipped; re-harvested sessions overwrite old summaries
|
|
151
|
+
|
|
152
|
+
**Quick setup with ai-proxy:**
|
|
153
|
+
|
|
154
|
+
```yaml
|
|
155
|
+
summarization:
|
|
156
|
+
enabled: true
|
|
157
|
+
provider_url: "https://ai-proxy.example.com/v1"
|
|
158
|
+
api_key: "" # set NANO_BRAIN_SUMMARIZE_API_KEY instead
|
|
159
|
+
model: "claude-sonnet-4-5"
|
|
160
|
+
max_tokens: 4096
|
|
161
|
+
concurrency: 3
|
|
162
|
+
output_dir: "~/.nano-brain/summaries"
|
|
133
163
|
```
|
|
134
164
|
|
|
165
|
+
Or via environment variable:
|
|
166
|
+
|
|
167
|
+
```bash
|
|
168
|
+
export NANO_BRAIN_SUMMARIZE_API_KEY="sk-..."
|
|
169
|
+
```
|
|
170
|
+
|
|
171
|
+
Large sessions (100K+ tokens) are handled via map-reduce chunking — no session is too large.
|
|
172
|
+
|
|
135
173
|
### Environment Variables
|
|
136
174
|
|
|
137
175
|
| Variable | Description |
|
|
@@ -139,6 +177,7 @@ logging:
|
|
|
139
177
|
| `DATABASE_URL` | PostgreSQL connection string |
|
|
140
178
|
| `VOYAGE_API_KEY` | Voyage AI API key |
|
|
141
179
|
| `OPENCODE_STORAGE_DIR` | OpenCode session directory |
|
|
180
|
+
| `NANO_BRAIN_SUMMARIZE_API_KEY` | API key for the summarization LLM provider |
|
|
142
181
|
| `NANO_BRAIN_*` | Override any config (e.g., `NANO_BRAIN_SERVER_PORT=3100`) |
|
|
143
182
|
|
|
144
183
|
## REST API
|