@nano-step/nano-brain 2026.5.2902 → 2026.5.2903
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 +12 -8
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -111,7 +111,9 @@ search:
|
|
|
111
111
|
|
|
112
112
|
harvester:
|
|
113
113
|
opencode:
|
|
114
|
-
|
|
114
|
+
db_root: "" # e.g., ~/.ai-sandbox/opencode-dbs (multi-DB, highest priority)
|
|
115
|
+
db_path: "" # e.g., ~/.local/share/opencode/opencode.db (single DB)
|
|
116
|
+
session_dir: "" # e.g., ~/.local/share/opencode/storage (legacy JSON)
|
|
115
117
|
claudecode:
|
|
116
118
|
enabled: false
|
|
117
119
|
session_dir: ""
|
|
@@ -136,19 +138,19 @@ summarization:
|
|
|
136
138
|
provider_url: "" # OpenAI-compatible endpoint, e.g. https://ai-proxy.example.com/v1
|
|
137
139
|
api_key: "" # or set NANO_BRAIN_SUMMARIZE_API_KEY env var
|
|
138
140
|
model: "nano-brain" # model name passed to the provider
|
|
139
|
-
max_tokens:
|
|
141
|
+
max_tokens: 8000 # max tokens per LLM completion
|
|
140
142
|
concurrency: 3 # parallel map-phase LLM calls
|
|
141
|
-
output_dir: "~/.nano-brain/summaries" # directory for .md summary files
|
|
142
143
|
```
|
|
143
144
|
|
|
144
145
|
### Session Summarization
|
|
145
146
|
|
|
146
147
|
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
|
|
|
148
|
-
-
|
|
149
|
-
- Stored in the vector DB under collection `session-summary` for semantic search
|
|
149
|
+
- Stored in PostgreSQL under collection `session-summary` for semantic search via the standard query/vsearch API (PG is the source of truth)
|
|
150
150
|
- Idempotent — unchanged sessions are skipped; re-harvested sessions overwrite old summaries
|
|
151
151
|
|
|
152
|
+
> **Note**: as of `harvest-summary-only` (May 2026), summaries are no longer written to disk as `.md` files. The legacy `output_dir` YAML key is silently ignored for backward compat. Any pre-existing files under `~/.nano-brain/summaries/` are stale artifacts and can be safely deleted.
|
|
153
|
+
|
|
152
154
|
**Quick setup with ai-proxy:**
|
|
153
155
|
|
|
154
156
|
```yaml
|
|
@@ -157,9 +159,8 @@ summarization:
|
|
|
157
159
|
provider_url: "https://ai-proxy.example.com/v1"
|
|
158
160
|
api_key: "" # set NANO_BRAIN_SUMMARIZE_API_KEY instead
|
|
159
161
|
model: "claude-sonnet-4-5"
|
|
160
|
-
max_tokens:
|
|
162
|
+
max_tokens: 8000
|
|
161
163
|
concurrency: 3
|
|
162
|
-
output_dir: "~/.nano-brain/summaries"
|
|
163
164
|
```
|
|
164
165
|
|
|
165
166
|
Or via environment variable:
|
|
@@ -176,7 +177,9 @@ Large sessions (100K+ tokens) are handled via map-reduce chunking — no session
|
|
|
176
177
|
|----------|-------------|
|
|
177
178
|
| `DATABASE_URL` | PostgreSQL connection string |
|
|
178
179
|
| `VOYAGE_API_KEY` | Voyage AI API key |
|
|
179
|
-
| `
|
|
180
|
+
| `OPENCODE_DB_ROOT` | OpenCode per-project DB root directory (multi-DB mode) |
|
|
181
|
+
| `OPENCODE_DB_PATH` | OpenCode single SQLite database path |
|
|
182
|
+
| `OPENCODE_STORAGE_DIR` | OpenCode session directory (legacy) |
|
|
180
183
|
| `NANO_BRAIN_SUMMARIZE_API_KEY` | API key for the summarization LLM provider |
|
|
181
184
|
| `NANO_BRAIN_*` | Override any config (e.g., `NANO_BRAIN_SERVER_PORT=3100`) |
|
|
182
185
|
|
|
@@ -234,6 +237,7 @@ Workspace is passed in the JSON body for POST, query param for GET.
|
|
|
234
237
|
| `nano-brain vsearch` | Vector similarity search |
|
|
235
238
|
| `nano-brain collection add\|remove\|list` | Manage collections |
|
|
236
239
|
| `nano-brain harvest` | Trigger session harvesting |
|
|
240
|
+
| `nano-brain cleanup-stale-raw [--dry-run]` | Delete pre-#192 raw OpenCode session docs superseded by summaries |
|
|
237
241
|
| `nano-brain bench generate\|run\|compare\|stress` | Benchmarking suite |
|
|
238
242
|
| `nano-brain db:migrate` | Run pending goose migrations |
|
|
239
243
|
| `nano-brain db:migrate --from-v1 <path>` | Import V1 SQLite data |
|