@goondocks/myco 0.5.0 → 0.5.1

package/.claude-plugin/marketplace.json

@@ -12,7 +12,7 @@
       "source": {
         "source": "npm",
         "package": "@goondocks/myco",
-        "version": "0.4.4"
+        "version": "0.5.0"
       },
       "description": "Collective agent intelligence — captures session knowledge and serves it back via MCP",
       "license": "MIT",

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "myco",
-  "version": "0.5.0",
+  "version": "0.5.1",
   "description": "Collective agent intelligence — captures session knowledge and serves it back to your team via MCP",
   "author": {
     "name": "goondocks-co",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@goondocks/myco",
-  "version": "0.5.0",
+  "version": "0.5.1",
   "description": "Collective agent intelligence — Claude Code plugin",
   "type": "module",
   "main": "dist/index.js",

package/skills/myco/SKILL.md

@@ -140,7 +140,7 @@ View daemon logs for debugging when sessions aren't being captured, observations
 { "level": "warn", "component": "processor", "limit": 20 }
 ```
 
-Components: `daemon`, `processor`, `hooks`, `lifecycle`, `embeddings`, `lineage`, `watcher`.
+Components: `daemon`, `processor`, `hooks`, `lifecycle`, `embeddings`, `lineage`, `watcher`, `digest`, `curation`.
 
 ### myco_supersede — Mark a spore as replaced
 
@@ -237,6 +237,25 @@ Options:
 - `--session <id>` — reprocess a single session (partial ID match)
 - `--index-only` — skip LLM extraction, just re-index and re-embed existing notes
 
+### Digest management
+
+```
+node <plugin-root>/dist/src/cli.js digest              # Run incremental digest cycle
+node <plugin-root>/dist/src/cli.js digest --tier 3000  # Reprocess a specific tier (clean slate)
+node <plugin-root>/dist/src/cli.js digest --full       # Reprocess all tiers from scratch
+```
+
+### Vault curation
+
+Supersession happens automatically on every spore write. For vault-wide cleanup, see `references/cli-usage.md` for full flags:
+
+```
+node <plugin-root>/dist/src/cli.js curate              # Scan and supersede stale spores
+node <plugin-root>/dist/src/cli.js curate --dry-run    # Preview without writing
+```
+
+For patterns on when to manually supersede or consolidate, see `references/wisdom.md`.
+
 ### Other maintenance commands
 
 ```

package/skills/myco/references/cli-usage.md

@@ -287,6 +287,54 @@ Run this after changing the embedding model (via `setup-llm`) to regenerate all
 
 ---
 
+### `digest` — Run a digest cycle on demand
+
+Trigger a digest cycle manually. Use `--tier` to reprocess a specific tier from scratch (all substrate, no previous extract), or `--full` for a complete rebuild of all tiers.
+
+| Flag | Type | Description |
+|------|------|-------------|
+| `--tier <number>` | number | Reprocess a specific tier (clean slate) |
+| `--full` | boolean | Reprocess all tiers from scratch |
+
+When `--tier` or `--full` is used, the cycle reads all vault notes (ignoring the last-cycle timestamp) and skips the previous extract, producing a fresh synthesis.
+
+**Examples:**
+
+```sh
+# Run an incremental cycle (same as what the metabolism timer does)
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js digest
+
+# Reprocess tier 3000 from scratch
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js digest --tier 3000
+
+# Full rebuild of all tiers
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js digest --full
+```
+
+---
+
+### `curate` — Scan vault and supersede stale spores
+
+Scans all active spores, clusters them by semantic similarity within each observation type, and asks the LLM which older spores are outdated. Superseded spores are preserved with lineage metadata — never deleted.
+
+| Flag | Type | Description |
+|------|------|-------------|
+| `--dry-run` | boolean | Run LLM evaluation but print results without writing |
+
+**Examples:**
+
+```sh
+# Scan and supersede stale spores
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js curate
+
+# Preview what would be superseded
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js curate --dry-run
+```
+
+Note: `--dry-run` still runs LLM calls (to evaluate clusters) — it just skips the writes. Use it to review before running on a vault for the first time.
+
+---
+
 ### `reprocess` — Re-extract observations from transcripts
 
 Re-reads session transcripts, re-extracts observations with the current LLM, and re-indexes. Existing spores are preserved — new extractions are additive.

package/skills/myco/references/wisdom.md

@@ -2,9 +2,19 @@
 
 When you notice patterns in vault spores — recurring themes, conflicting advice, outdated observations — use these tools to keep the vault clean and its knowledge sharp.
 
+## Automatic Curation
+
+Myco automatically checks for supersession every time a new spore is written. After the spore is saved and embedded, a fire-and-forget pipeline searches for semantically similar active spores of the same observation type and asks the LLM whether any are now outdated. If so, they're marked superseded automatically. This means most vault hygiene happens without manual intervention.
+
+For vault-wide cleanup (e.g., after a large refactor), use the CLI:
+```sh
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js curate --dry-run  # preview
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js curate             # execute
+```
+
 ## Supersede
 
-Use `myco_supersede` when a newer spore replaces an older one.
+Use `myco_supersede` for manual supersession when you spot a stale spore that automatic curation missed.
 
 **Signals:**
 - A decision was reversed in a later session