@goondocks/myco 0.3.7 → 0.4.1

package/.claude-plugin/marketplace.json

@@ -12,7 +12,7 @@
       "source": {
         "source": "npm",
         "package": "@goondocks/myco",
-        "version": "0.3.6"
+        "version": "0.4.0"
       },
       "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": "0.3.7",
+  "version": "0.4.1",
   "description": "Collective agent intelligence — captures session knowledge and serves it back to your team via MCP",
   "author": {
     "name": "goondocks-co",

package/README.md

@@ -37,9 +37,10 @@ Myco captures everything your AI agents do — sessions, decisions, plans, disco
 myco_search("how did we handle auth?")  → semantically matched sessions, decisions, and linked context
 myco_recall("migration plan")           → full decision history with session lineage
 myco_remember(observation)              → persist a discovery for the team
+myco_context(tier: 3000)                → pre-computed project understanding, instantly available
 ```
 
-**For humans** — open the vault in Obsidian and browse the intelligence graph visually. Sessions link to plans, plans link to decisions, decisions link to memories. It's all Markdown with backlinks — your team's connected knowledge, navigable and searchable.
+**For humans** — open the vault in Obsidian and browse the intelligence graph visually. Sessions link to plans, plans link to decisions, decisions link to spores. It's all Markdown with backlinks — your team's connected knowledge, navigable and searchable.
 
 **For teams** — the vault is a Git-friendly directory of Markdown files. Share it through your existing Git workflow.
 
@@ -47,7 +48,11 @@ myco_remember(observation)              → persist a discovery for the team
 
 ### Capture
 
-A background daemon reads your agent's conversation transcript after each turn — the full dialogue including prompts, AI responses, tool calls, and screenshots. Observations (decisions, gotchas, discoveries) are extracted automatically via a local LLM and written as linked vault notes.
+A background daemon reads your agent's conversation transcript after each turn — the full dialogue including prompts, AI responses, tool calls, and screenshots. Observations called **spores** (decisions, gotchas, discoveries, trade-offs, bug fixes) are extracted automatically via a local LLM and written as linked vault notes.
+
+### Digest
+
+A **continuous reasoning engine** runs inside the daemon, periodically synthesizing all accumulated knowledge into tiered context extracts. These pre-computed summaries give agents an instant, rich understanding of the project at session start — no searching required. Four tiers serve different needs: executive briefing (1.5K tokens), team standup (3K), deep onboarding (5K), and institutional knowledge (10K).
 
 ### Index
 
@@ -55,11 +60,11 @@ Every note is indexed for both keyword search (SQLite FTS5) and semantic search
 
 ### Serve
 
-An MCP server exposes the vault to any agent runtime. Relevant memories are injected into every prompt automatically — no manual lookup needed. Agents build on your team's accumulated knowledge without being told to.
+An MCP server exposes the vault to any agent runtime. The digest extract is injected at session start for immediate context, and relevant spores are injected per-prompt for targeted intelligence. Agents build on your team's accumulated knowledge without being told to.
 
 ### Connect
 
-Sessions link to plans. Plans link to decisions. Decisions link to memories. Obsidian backlinks and metadata create a navigable graph of your team's institutional knowledge. Open the vault in [Obsidian](https://obsidian.md) to browse it visually, or let agents traverse it via MCP tools.
+Sessions link to plans. Plans link to decisions. Decisions link to spores. Obsidian backlinks and metadata create a navigable graph of your team's institutional knowledge. Open the vault in [Obsidian](https://obsidian.md) to browse it visually, or let agents traverse it via MCP tools.
 
 ### Multi-agent
 

package/commands/init.md

@@ -9,20 +9,24 @@ Guide the user through setup using the composable CLI commands. **Do NOT create
 
 **Ask each question one at a time using AskUserQuestion with selectable options.** Wait for the user's answer before proceeding to the next question. Do NOT combine multiple questions into one message.
 
-## Step 1: Detect available providers
+The streamlined setup asks just four questions: vault location, provider, model, and embedding model. One model handles everything — hooks, extraction, summaries, and digest — sized for the most demanding task (digestion). Advanced configuration is available via CLI commands after init.
 
-Run the provider detection command to see what's available:
+## Step 1: Detect available providers and system capabilities
+
+Run the provider detection command and detect system RAM:
 
 ```bash
 node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js detect-providers
 ```
 
+Detect RAM:
+- **macOS**: `sysctl -n hw.memsize` (bytes → GB)
+- **Linux**: parse `/proc/meminfo` for `MemTotal`
+
 Parse the JSON output. This tells you which providers are running and what models are available.
 
 ## Step 2: Choose vault location
 
-Ask the user:
-
 **Question:** "Where would you like to store the Myco vault?"
 
 **Options:**
@@ -30,69 +34,89 @@ Ask the user:
 - "Centralized (~/.myco/vaults/<project-name>/)" — vault stays outside the repo, good for public repos or personal use
 - "Custom path" — specify your own location
 
-If the user picks "Custom path", ask them to type the path.
+## Step 3: Choose provider and model
+
+**Question:** "Which LLM provider and model?"
+
+List only providers where `available` is `true`. Recommend a model sized for digest based on detected RAM:
+
+| RAM | Recommended Model | Digest Context |
+|-----|-------------------|----------------|
+| **64GB+** | `qwen3.5:35b` (MoE, recommended) | 65536 |
+| **32–64GB** | `qwen3.5:27b` | 32768 |
+| **16–32GB** | `qwen3.5:latest` (~10B) | 16384 |
+| **8–16GB** | `qwen3.5:4b` | 8192 |
 
-## Step 3: Choose LLM provider
+The same model handles hooks (at 8K context), extraction, summaries, and digest (at the larger context from the table). No separate model configuration needed.
 
-Using the detected providers from Step 1, ask the user:
+If the model isn't installed, offer to pull it:
+- **Ollama**: `ollama pull qwen3.5`
+- **LM Studio**: search for `qwen3.5` in the model browser
 
-**Question:** "Which LLM provider for summarization?"
+## Step 4: Choose embedding model
 
-**Options:** List only providers where `available` is `true`, with recommended models. Example:
-- "Ollama — gpt-oss (recommended)"
-- "LM Studio — openai/gpt-oss-20b"
-- "Anthropic"
+**Question:** "Which embedding model?"
 
-After the user picks a provider, ask them to choose a specific model from that provider's model list (from the detect-providers output).
+**Options:** List only providers that support embeddings (Anthropic does not):
+- **Ollama** — list available embedding models. If none are available, offer to pull one (e.g., `bge-m3` or `nomic-embed-text`).
+- **LM Studio** — filter the model list for names containing `text-embedding`. If none are available, guide the user to search for and download an embedding model through LM Studio's model browser.
 
-## Step 4: Choose embedding provider
+If no embedding models are available on the chosen provider, help the user get one before proceeding.
 
-Ask the user:
+## Step 5: Choose digest inject tier
 
-**Question:** "Which embedding provider?"
+**Question:** "How much context should the agent receive at session start?"
 
-**Options:** List only providers where `available` is `true` and that support embeddings (Anthropic does not). Example:
-- "Ollama — bge-m3 (recommended)"
-- "LM Studio — text-embedding-bge-m3"
+Based on RAM, present the recommended tiers:
 
-After the user picks a provider, ask them to choose a specific embedding model.
+| RAM | Options | Default |
+|-----|---------|---------|
+| **64GB+** | 1500, 3000, 5000, 10000 | 3000 |
+| **32–64GB** | 1500, 3000, 5000 | 3000 |
+| **16–32GB** | 1500, 3000 | 1500 |
+| **8–16GB** | 1500 | 1500 |
 
-If the recommended embedding model isn't available, offer to pull it:
-- **Ollama**: `ollama pull bge-m3`
+**Options:**
+- "1500 — executive briefing (fastest, lightest)"
+- "3000 — team standup (recommended)"
+- "5000 — deep onboarding"
+- "10000 — institutional knowledge (richest)"
+
+This controls what gets auto-injected at the start of every session. Agents can always request a different tier on-demand via the `myco_context` tool.
 
-## Step 5: Run init with all gathered inputs
+## Step 6: Run init and configure
 
-Pass everything to the init command in a single call:
+Create the vault and apply settings:
 
 ```bash
+# Create vault structure and base config
 node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js init \
   --vault <chosen-path> \
   --llm-provider <provider> \
   --llm-model <model> \
-  --llm-url <base-url> \
-  --embedding-provider <provider> \
-  --embedding-model <model> \
-  --embedding-url <base-url>
-```
+  --embedding-provider <embedding-provider> \
+  --embedding-model <embedding-model>
 
-The CLI creates the vault structure, writes myco.yaml, .gitignore, _dashboard.md, initializes the FTS index, and configures MYCO_VAULT_DIR if the vault is external.
-
-## Step 6: Verify connectivity
+# Set digest context window and inject tier based on user choices
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-digest \
+  --context-window <from-ram-table> \
+  --inject-tier <chosen-tier>
+```
 
-Run the verify command to confirm providers are reachable:
+## Step 7: Verify connectivity
 
 ```bash
 node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js verify
 ```
 
-If verification fails, help the user troubleshoot (check if the provider is running, model is loaded, etc.).
-
-## Step 7: Display summary
+If verification fails, help the user troubleshoot.
 
-Show the user a setup summary table:
+## Step 8: Display summary
 
 | Setting | Value |
 |---------|-------|
 | Vault path | `<resolved path>` |
-| LLM provider | `<provider>` / `<model>` |
-| Embedding provider | `<provider>` / `<model>` |
+| Provider | `<provider>` / `<model>` |
+| Embedding | `<embedding-provider>` / `<embedding-model>` |
+| Digest | enabled (context: `<context-window>`) |
+| RAM detected | `<X>` GB |

package/commands/setup-llm.md

@@ -7,11 +7,13 @@ description: Configure or change the intelligence backend (Ollama, LM Studio, or
 
 Guide the user through configuring their intelligence backend. This command can be run at any time to change providers or models.
 
+The streamlined setup asks just three questions: provider, model, and embedding model. One model handles everything — hooks, extraction, summaries, and digest — at different context windows per request. Advanced configuration is available via the CLI for power users.
+
 ## Prerequisites
 
 Read the existing `myco.yaml` from the vault directory to show current settings before making changes.
 
-## Step 1: Detect available providers
+## Step 1: Detect available providers and system capabilities
 
 Check which providers are reachable:
 
@@ -19,66 +21,62 @@ Check which providers are reachable:
 - **LM Studio** — fetch `http://localhost:1234/v1/models`, list model names
 - **Anthropic** — check if `ANTHROPIC_API_KEY` is set in the environment
 
-Report which are available and which are not.
-
-## Step 2: Choose LLM provider
+Detect system RAM for recommendations:
+- **macOS**: `sysctl -n hw.memsize` (bytes → GB)
+- **Linux**: parse `/proc/meminfo` for `MemTotal`
 
-Ask the user to select from available providers:
+Report which providers are available and the detected RAM.
 
-- **Ollama** — list available models
-- **LM Studio** — list available models
-- **Anthropic** — verify API key works, default model `claude-haiku-4-5-20251001`
+## Step 2: Choose provider and model
 
-Recommended summarization models by hardware tier:
+Ask the user to select from available providers. After picking a provider, recommend a model sized for digest (the most demanding task). The same model handles hooks and extraction at smaller context windows automatically.
 
-| Tier | Models | RAM |
-|------|--------|-----|
-| **High** | `gpt-oss` (~20B), `gemma3:27b`, `qwen3.5:14b` | 16GB+ |
-| **Mid** | `qwen3.5:8b`, `gemma3:12b` | 8GB+ |
-| **Light** | `gemma3:4b`, `qwen3.5:4b` | 4GB+ |
+Recommended models by hardware tier — Qwen 3.5 is preferred for its strong instruction-following and synthesis quality:
 
-Any instruction-tuned model that handles JSON output works. Prefer what the user already has loaded.
+| RAM | Model | Context for Digest |
+|-----|-------|--------------------|
+| **64GB+** | `qwen3.5:35b` (MoE, recommended) | 65536 |
+| **32–64GB** | `qwen3.5:27b` | 32768 |
+| **16–32GB** | `qwen3.5:latest` (~10B) | 16384 |
+| **8–16GB** | `qwen3.5:4b` | 8192 |
 
-For local providers (Ollama, LM Studio), also configure:
-- `context_window` — ask or accept default of 8192
-- `max_tokens` — ask or accept default of 1024
+Any instruction-tuned model that handles JSON output works. Prefer what the user already has loaded, but recommend Qwen 3.5 if they're starting fresh.
 
 If the chosen model isn't installed, offer to pull it:
-- **Ollama**: `ollama pull gpt-oss` (pulls latest tag automatically)
-- **LM Studio**: `lms get openai/gpt-oss-20b` (uses `owner/model` format)
+- **Ollama**: `ollama pull qwen3.5` (pulls latest tag automatically)
+- **LM Studio**: search for `qwen3.5` in the model browser
 
-These settings do not apply to Anthropic (API-managed).
+## Step 3: Choose embedding model
 
-## Step 3: Choose embedding provider
+Ask the user to select an embedding model — **Anthropic is not an option** (it doesn't support embeddings):
 
-Ask the user to select from available providers — **Anthropic is not an option** (it doesn't support embeddings):
+- **Ollama** — list available embedding models. If none are available, offer to pull one (e.g., `bge-m3` or `nomic-embed-text`).
+- **LM Studio** — filter the model list for names containing `text-embedding`. If none are available, guide the user to search for and download an embedding model through LM Studio's model browser.
 
-- **Ollama** (recommended for embeddings) — list available models, recommend **`bge-m3`** or `nomic-embed-text`
-- **LM Studio** — possible but not recommended for embeddings; better suited for LLM work
+If no embedding models are available on the chosen provider, help the user get one before proceeding.
 
-If the embedding model isn't installed: `ollama pull bge-m3`
-
-**Important:** If the user changes the embedding model, the vector index must be rebuilt. Warn them:
+**Important:** If the user changes the embedding model, warn them:
 > "Changing the embedding model will require a full rebuild of the vector index. Run `node dist/src/cli.js rebuild` after this change."
 
-## Step 4: Update `myco.yaml`
-
-Write both `intelligence.llm` and `intelligence.embedding` sections with all values explicit:
-
-```yaml
-intelligence:
-  llm:
-    provider: ollama
-    model: gpt-oss
-    base_url: http://localhost:11434
-    context_window: 8192
-    max_tokens: 1024
-  embedding:
-    provider: ollama
-    model: bge-m3
-    base_url: http://localhost:11434
+## Step 4: Apply settings
+
+Use the CLI commands to write settings deterministically. The context window for the main LLM stays at 8192 (hooks don't need more). The digest context window is set based on the RAM tier recommendation.
+
+```bash
+# Set provider and model
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-llm \
+  --llm-provider <provider> \
+  --llm-model <model> \
+  --embedding-provider <embedding-provider> \
+  --embedding-model <embedding-model>
+
+# Set digest context window based on RAM tier (model inherits from main LLM)
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-digest \
+  --context-window <from-ram-table>
 ```
 
+Only pass flags the user explicitly changed — Zod defaults handle the rest.
+
 If migrating from a v1 config (has `backend: local/cloud` structure), bump `version` to `2` and rewrite the entire intelligence section. The loader auto-maps `provider: haiku` to `anthropic`.
 
 ## Step 5: Verify and restart
@@ -87,3 +85,30 @@ If migrating from a v1 config (has `backend: local/cloud` structure), bump `vers
 2. Test the embedding provider with a test embedding
 3. Restart the daemon to pick up the new config: `node dist/src/cli.js restart`
 4. Report success or issues found
+
+## Advanced Configuration
+
+For power users who want fine-grained control, all settings are available via CLI:
+
+```bash
+# Separate digest model (e.g., larger model on LM Studio)
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-digest \
+  --provider lm-studio \
+  --model "qwen/qwen3.5-35b-a3b" \
+  --context-window 65536 \
+  --gpu-kv-cache false
+
+# Custom tiers and injection
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-digest \
+  --tiers 1500,3000,5000,10000 \
+  --inject-tier 3000
+
+# Capture token budgets
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-digest \
+  --extraction-tokens 2048 \
+  --summary-tokens 1024
+
+# View current settings
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-llm --show
+node ${CLAUDE_PLUGIN_ROOT}/dist/src/cli.js setup-digest --show
+```

package/commands/status.md

@@ -39,14 +39,24 @@ Query the FTS index for counts:
 | Metric | How to check |
 |--------|-------------|
 | Sessions | `index.query({ type: 'session' }).length` |
-| Memories | `index.query({ type: 'memory' }).length` |
+| Spores | `index.query({ type: 'spore' }).length` |
 | Plans | `index.query({ type: 'plan' }).length` |
 | Artifacts | `index.query({ type: 'artifact' }).length` |
 | Embeddings | Vector index count |
 
-Also report memory breakdown by observation type (decision, gotcha, trade_off, etc.).
+Also report spore breakdown by observation type (decision, gotcha, trade_off, etc.).
 
-## Step 5: Intelligence backend health
+## Step 5: Digest status
+
+Check the digest system state:
+
+- **Enabled/disabled**: read `digest.enabled` from `myco.yaml`
+- **Extracts**: list which tier files exist in `vault/digest/` (extract-1500.md, etc.) with file sizes and generated timestamps
+- **Last cycle**: read last line of `vault/digest/trace.jsonl` — report cycle ID, timestamp, tiers generated, substrate count, duration
+- **Metabolism**: report configured tiers, inject tier, and context window
+- **Digest model**: if `digest.intelligence.model` is set, show it; otherwise note "inherits from main LLM"
+
+## Step 6: Intelligence backend health
 
 Test connectivity to the configured providers:
 
@@ -54,7 +64,7 @@ Test connectivity to the configured providers:
 - **Embedding provider**: call `isAvailable()` — report reachable or not
 - If either is unreachable, suggest running `/myco-setup-llm`
 
-## Step 6: Pending issues
+## Step 7: Pending issues
 
 Check for problems:
 
@@ -63,13 +73,13 @@ Check for problems:
 - **Missing vectors**: does `vectors.db` exist? If not, embeddings are disabled
 - **Lineage**: does `lineage.json` exist? Report link count if so
 
-## Step 7: Recent activity
+## Step 8: Recent activity
 
 Show the 3 most recent sessions with:
 - Session ID (short form)
 - Title
 - Started/ended timestamps
-- Number of memories extracted
+- Number of spores extracted
 - Parent session (if lineage detected)
 
 ## Output format
@@ -92,18 +102,26 @@ Sessions: 1 active
 
 --- Vault ---
 Sessions:  12
-Memories:  183 (67 decision, 34 gotcha, 32 trade_off, 20 discovery, 19 bug_fix, 1 cross-cutting)
+Spores:    183 (67 decision, 34 gotcha, 32 trade_off, 20 discovery, 19 bug_fix, 1 cross-cutting)
 Plans:     0
 Artifacts: 8
 Vectors:   224
 
+--- Digest ---
+Enabled:    yes
+Tiers:      [1500, 3000, 5000, 10000]
+Inject:     3000 (auto-inject at session start)
+Model:      gpt-oss (inherited from main LLM)
+Last cycle: dc-a1b2c3 (2 min ago, 4 tiers, 12 notes, 45s)
+Extracts:   1500 (1.1KB), 3000 (4.5KB), 5000 (6.9KB), 10000 (9.6KB)
+
 --- Lineage ---
 Links: 5 (3 clear, 1 inferred, 1 semantic_similarity)
 
 --- Recent Sessions ---
-1. [abc123] "Auth redesign session" (2h 15m, 5 memories)
-2. [def456] "Bug fix for CORS" (45m, 2 memories, parent: abc123)
-3. [ghi789] "Config cleanup" (20m, 1 memory)
+1. [abc123] "Auth redesign session" (2h 15m, 5 spores)
+2. [def456] "Bug fix for CORS" (45m, 2 spores, parent: abc123)
+3. [ghi789] "Config cleanup" (20m, 1 spore)
 
 --- Issues ---
 None found.

package/dist/{chunk-YFG2O5HR.js → chunk-2GJFTIWX.js}

@@ -1,7 +1,7 @@
 import { createRequire as __cr } from 'node:module'; const require = __cr(import.meta.url);
 import {
   AgentRegistry
-} from "./chunk-JKOALBZC.js";
+} from "./chunk-BNIYWCST.js";
 
 // src/version.ts
 import fs from "fs";
@@ -30,4 +30,4 @@ function readVersionFrom(dir) {
 export {
   getPluginVersion
 };
-//# sourceMappingURL=chunk-YFG2O5HR.js.map
+//# sourceMappingURL=chunk-2GJFTIWX.js.map