@zilliz/memsearch-opencode 0.1.0 → 0.1.2

package/README.md

@@ -21,11 +21,20 @@ uv tool install 'memsearch[onnx]'
 # or: pip install 'memsearch[onnx]'
 ```
 
-### Install
+### Install from npm (recommended)
+
+```json
+// In ~/.config/opencode/opencode.json
+{
+  "plugin": ["@zilliz/memsearch-opencode"]
+}
+```
+
+### Install from Source (development)
 
 ```bash
-# Clone the repo (if you haven't already)
-git clone https://github.com/zc277584121/memsearch.git
+# Clone the repo
+git clone https://github.com/zilliztech/memsearch.git
 cd memsearch
 
 # Run the installer
@@ -44,15 +53,6 @@ mkdir -p ~/.agents/skills
 ln -sf /path/to/memsearch/plugins/opencode/skills/memory-recall ~/.agents/skills/memory-recall
 ```
 
-### npm Install (coming soon)
-
-```json
-// In ~/.config/opencode/opencode.json
-{
-  "plugin": ["memsearch-opencode"]
-}
-```
-
 ## Architecture
 
 ```
@@ -73,6 +73,20 @@ OpenCode Session
         └── memory_transcript ──→ parse-transcript.py (SQLite reader)
 ```
 
+## Recall Memories
+
+**Manual invocation** — explicitly invoke the skill with a query:
+
+```
+/memory-recall what was the auth approach we discussed?
+```
+
+**Auto invocation** — just ask naturally, the LLM auto-invokes memory tools when it senses the question needs history:
+
+```
+We discussed the authentication flow before, what was the approach?
+```
+
 ## Tools
 
 | Tool | Description |
@@ -100,7 +114,7 @@ Each file contains timestamped entries with bullet-point summaries:
 ## Session 14:30
 
 ### 14:30
-<!-- session:ses_abc123 source:opencode-sqlite -->
+<!-- session:ses_abc123 db:~/.local/share/opencode/opencode.db -->
 - User asked about the authentication flow.
 - Assistant explained the OAuth2 implementation in auth.ts.
 - Assistant modified the token refresh logic in refresh.ts.

package/index.ts

@@ -68,12 +68,15 @@ function deriveCollectionName(projectDir: string): string {
 }
 
 /**
- * Read the tail of the N most recent daily .md files for cold-start context.
+ * Summarize the N most recent daily .md files for cold-start context.
+ * Extracts headings (## Session, ### turns) and bullet content from each
+ * file so the agent sees the structure of past days (what sessions existed,
+ * what topics came up), not just the tail of whichever file is newest.
  */
 function getRecentMemories(
   memDir: string,
   count = 2,
-  tailLines = 15
+  maxLinesPerFile = 30
 ): string {
   if (!existsSync(memDir)) return "";
 
@@ -84,23 +87,24 @@ function getRecentMemories(
 
   if (files.length === 0) return "";
 
-  const bullets: string[] = [];
+  const summary: string[] = [];
   for (const file of files) {
     try {
       const content = readFileSync(join(memDir, file), "utf-8");
-      const lines = content.split("\n").slice(-tailLines);
-      const fileBullets = lines.filter((l) => l.startsWith("- ") || l.startsWith("[Human]") || l.startsWith("[Assistant]"));
-      if (fileBullets.length > 0) {
-        bullets.push(`[${file}]`, ...fileBullets);
+      const lines = content.split("\n")
+        .filter((l) => /^#{2,4}\s/.test(l) || l.startsWith("- ") || l.startsWith("[Human]") || l.startsWith("[Assistant]"))
+        .slice(0, maxLinesPerFile);
+      if (lines.length > 0) {
+        summary.push(`[${file}]`, ...lines);
       }
     } catch { /* skip */ }
   }
 
-  if (bullets.length === 0) {
+  if (summary.length === 0) {
     return `You have ${files.length} past memory file(s). Use the memory_search tool when the user's question could benefit from historical context.`;
   }
 
-  return `Recent memories (use memory_search for full search):\n${bullets.join("\n")}`;
+  return `Recent memories (use memory_search for full search):\n${summary.join("\n")}`;
 }
 
 /** Shell-escape a string for safe use inside single quotes. */
@@ -281,7 +285,7 @@ const MemsearchPlugin: Plugin = async ({ project, directory, worktree }) => {
         description:
           "Retrieve the original conversation from a past OpenCode session. " +
           "Use after memory_get when the expanded result contains a session anchor " +
-          "(<!-- session:ID source:opencode-sqlite -->). Returns the formatted " +
+          "(<!-- session:ID db:PATH -->). Returns the formatted " +
           "dialogue with [Human] and [Assistant] labels.",
         args: {
           session_id: tool.schema.string().describe("The session ID from the anchor comment"),

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@zilliz/memsearch-opencode",
-  "version": "0.1.0",
+  "version": "0.1.2",
   "description": "memsearch plugin for OpenCode — semantic memory search across sessions",
   "type": "module",
   "main": "index.ts",

package/scripts/capture-daemon.py

@@ -208,7 +208,7 @@ def _extract_msg_text(conn, msg_id, msg_json):
     return "\n".join(text_parts)
 
 
-def write_capture(memory_dir, turn_text, session_id):
+def write_capture(memory_dir, turn_text, session_id, db_path=""):
     """Write a captured turn to the daily memory file."""
     os.makedirs(memory_dir, exist_ok=True)
 
@@ -220,7 +220,7 @@ def write_capture(memory_dir, turn_text, session_id):
         with open(memory_file, "w") as f:
             f.write(f"# {today}\n\n## Session {now}\n\n")
 
-    anchor = f"<!-- session:{session_id} source:opencode-sqlite -->\n" if session_id else ""
+    anchor = f"<!-- session:{session_id} db:{db_path} -->\n" if session_id else ""
     entry = f"### {now}\n{anchor}{turn_text}\n\n"
 
     with open(memory_file, "a") as f:
@@ -280,7 +280,7 @@ def main():
                 if turn_text and len(turn_text) > 10:
                     # Summarize with LLM, fallback to raw text
                     summary = summarize_with_llm(turn_text, small_model)
-                    write_capture(memory_dir, summary if summary else turn_text, session_id)
+                    write_capture(memory_dir, summary if summary else turn_text, session_id, db_path)
                     if msg_time > last_msg_time:
                         last_msg_time = msg_time
                         try:

package/skills/memory-recall/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: memory-recall
-description: "Search and recall relevant memories from past sessions. Use when the user's question could benefit from historical context, past decisions, debugging notes, previous conversations, or project knowledge."
+description: "Search and recall relevant memories from past sessions via memsearch. Use when the user's question could benefit from historical context, past decisions, debugging notes, previous conversations, or project knowledge -- especially questions like 'what did I decide about X', 'why did we do Y', or 'have I seen this before'. Also use when you see `[memsearch] Memory available` hints injected via SessionStart or UserPromptSubmit. Typical flow: search for 3-5 chunks, expand the most relevant, optionally deep-drill into original transcripts via the anchor format. Skip when the question is purely about current code state (use Read/Grep), ephemeral (today's task only), or the user has explicitly asked to ignore memory."
 allowed-tools: Bash
 ---
 
@@ -8,7 +8,7 @@ You are a memory retrieval agent for memsearch. Your job is to search past memor
 
 ## Project Collection
 
-Collection: !`bash __INSTALL_DIR__/scripts/derive-collection.sh`
+Collection: !`bash -c 'root=$(git rev-parse --show-toplevel 2>/dev/null || true); if [ -n "$root" ]; then bash __INSTALL_DIR__/scripts/derive-collection.sh "$root"; else bash __INSTALL_DIR__/scripts/derive-collection.sh; fi'`
 
 ## Your Task
 
@@ -24,14 +24,22 @@ Search for memories relevant to: $ARGUMENTS
 
 3. **Expand**: For each relevant result, run `memsearch expand <chunk_hash> --collection <collection name above>` to get the full markdown section with surrounding context.
 
-4. **Deep drill (optional)**: If an expanded chunk contains a session anchor (`<!-- session:ID source:opencode-sqlite -->`), and the original conversation seems critical, run:
-   ```
-   python3 __INSTALL_DIR__/scripts/parse-transcript.py <session_id> --limit 10
-   ```
-   to retrieve the original conversation turns from the OpenCode SQLite database.
+4. **Deep drill (optional)**: If an expanded chunk contains transcript anchors (HTML comments with session info), and the original conversation seems critical:
+   - If the anchor contains `db:`, run `python3 __INSTALL_DIR__/scripts/parse-transcript.py <session_id> --limit 10` to retrieve the original conversation turns from the SQLite database.
+   - If the anchor format is unfamiliar (e.g. `transcript:`, `rollout:` instead of `db:`), try reading the referenced file directly to explore its structure and locate the relevant conversation by the session or turn identifiers in the anchor.
 
 5. **Return results**: Output a curated summary of the most relevant memories. Be concise — only include information that is genuinely useful for the user's current question.
 
+## When unsure what to search
+
+If the user's question is vague or you can't form a concrete search query, explore the raw markdown first — it is the source of truth for memory:
+
+- `ls -t .memsearch/memory/ | head -10` — recent daily logs
+- `grep -h "^## " .memsearch/memory/*.md | sort -u | tail -40` — session headings across all days
+- `cat .memsearch/memory/<YYYY-MM-DD>.md` — read a specific day
+
+Once a concrete topic jumps out, go back to `memsearch search` with a specific query.
+
 ## Output Format
 
 Organize by relevance. For each memory include: