memindex 1.2.0 → 1.4.0
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/.claude-plugin/marketplace.json +1 -1
- package/.claude-plugin/plugin.json +1 -1
- package/.codex-plugin/plugin.json +1 -1
- package/.qoder-plugin/plugin.json +1 -1
- package/README.md +85 -41
- package/THIRD_PARTY_LICENSES +30 -0
- package/dist/bug-report/index.js +63 -63
- package/dist/npx-cli/index.js +417 -317
- package/dist/opencode-plugin/index.js +34 -34
- package/dist/pi-extension/index.js +6 -6
- package/integrations/dsh/lib/index.js +24 -24
- package/integrations/dsh/package.json +1 -1
- package/integrations/hermes/memindex/client.py +11 -6
- package/integrations/omp/hooks/memindex.ts +5 -1
- package/integrations/openclaw/TESTING.md +1 -1
- package/integrations/openclaw/dist/index.js +9 -9
- package/integrations/openclaw/e2e-verify.sh +8 -1
- package/integrations/openclaw/openclaw.plugin.json +1 -1
- package/integrations/openclaw/src/index.test.ts +3 -1
- package/integrations/openclaw/src/index.ts +15 -5
- package/package.json +2 -1
- package/plugin/.claude-plugin/plugin.json +1 -1
- package/plugin/.codex-plugin/plugin.json +1 -1
- package/plugin/.qoder-plugin/plugin.json +1 -1
- package/plugin/hooks.json +1 -13
- package/plugin/package.json +1 -1
- package/plugin/scripts/context-generator.cjs +72 -60
- package/plugin/scripts/context-generator.cjs.map +3 -3
- package/plugin/scripts/mcp-server.cjs +68 -68
- package/plugin/scripts/mcp-server.cjs.map +3 -3
- package/plugin/scripts/statusline.cjs +1 -0
- package/plugin/scripts/transcript-watcher.cjs +35 -35
- package/plugin/scripts/transcript-watcher.cjs.map +3 -3
- package/plugin/scripts/worker-service.cjs +655 -595
- package/plugin/scripts/worker-service.cjs.map +3 -3
- package/plugin/skills/cloud-sync/SKILL.md +5 -2
- package/plugin/skills/mode-creator/SKILL.md +1 -1
- package/plugin/skills/timeline-report/SKILL.md +8 -5
- package/plugin/skills/weekly-digests/SKILL.md +5 -2
- package/plugin/sqlite/SessionStore.js +247 -229
- package/plugin/sqlite/SessionStore.js.map +3 -3
- package/plugin/ui/tv.html +55 -28
- package/plugin/ui/viewer-bundle.js +22 -15
- package/plugin/ui/viewer.html +72 -0
|
@@ -34,7 +34,10 @@ Resolve the worker port and query the always-registered status route:
|
|
|
34
34
|
|
|
35
35
|
```bash
|
|
36
36
|
PORT="${MEMINDEX_WORKER_PORT:-$(node -e "const fs=require('fs'),p=require('path'),os=require('os');const uid=(typeof process.getuid==='function'?process.getuid():77);const fallback=String(38700+(uid%100));try{const s=JSON.parse(fs.readFileSync(p.join(os.homedir(),'.memindex','settings.json'),'utf-8'));process.stdout.write(String(s.MEMINDEX_WORKER_PORT||fallback));}catch{process.stdout.write(fallback);}" 2>/dev/null)}"
|
|
37
|
-
|
|
37
|
+
# memcurl sends the install secret without putting it in the command line
|
|
38
|
+
memcurl() { printf 'X-Memindex-Secret: %s
|
|
39
|
+
' "$(cat "${MEMINDEX_DATA_DIR:-$HOME/.memindex}/worker-secret")" | curl -H @- "$@"; }
|
|
40
|
+
memcurl -s "http://127.0.0.1:${PORT}/api/sync/status"
|
|
38
41
|
```
|
|
39
42
|
|
|
40
43
|
- `configured: true` and `hub.reachable: true` → the worker completed an
|
|
@@ -69,7 +72,7 @@ Once the settings are saved, poll the status route every five seconds for up to
|
|
|
69
72
|
30 seconds. If the worker was not restarted, restart it first:
|
|
70
73
|
|
|
71
74
|
```bash
|
|
72
|
-
|
|
75
|
+
memcurl -s -X POST "http://127.0.0.1:${PORT}/api/admin/restart"
|
|
73
76
|
```
|
|
74
77
|
|
|
75
78
|
Success means `configured: true`, `hub.reachable: true`, and `lastError: null`.
|
|
@@ -105,7 +105,7 @@ Verify all of the following:
|
|
|
105
105
|
1. Restart reports a new healthy worker and exits successfully.
|
|
106
106
|
2. The installed file exists under the resolved data directory.
|
|
107
107
|
3. `settings.json` names the intended `MEMINDEX_MODE` without displaying secrets.
|
|
108
|
-
4. Request full startup context with the `session_start_context` MCP tool when available. Otherwise call `/api/context/inject?project=mode-creator-verification&full=true` on the configured local worker.
|
|
108
|
+
4. Request full startup context with the `session_start_context` MCP tool when available. Otherwise call `/api/context/inject?project=mode-creator-verification&full=true` on the configured local worker, sending the `X-Memindex-Secret` header (the contents of `worker-secret` in the data directory).
|
|
109
109
|
5. Startup context contains `Mode: <mode name> (<mode id>)`.
|
|
110
110
|
|
|
111
111
|
If the worker falls back to `code`, inspect the worker log for a mode validation or lookup error, repair the mode, and repeat the restart. Do not describe a fallback as successful activation.
|
|
@@ -26,6 +26,9 @@ The memindex worker must be running. The project must have memindex observations
|
|
|
26
26
|
|
|
27
27
|
```bash
|
|
28
28
|
WORKER_PORT="${MEMINDEX_WORKER_PORT:-$(node -e "const fs=require('fs'),p=require('path'),os=require('os');const uid=(typeof process.getuid==='function'?process.getuid():77);const fallback=String(38700+(uid%100));try{const s=JSON.parse(fs.readFileSync(p.join(os.homedir(),'.memindex','settings.json'),'utf-8'));process.stdout.write(String(s.MEMINDEX_WORKER_PORT||fallback));}catch{process.stdout.write(fallback);}" 2>/dev/null)}"
|
|
29
|
+
# memcurl sends the install secret without putting it in the command line
|
|
30
|
+
memcurl() { printf 'X-Memindex-Secret: %s
|
|
31
|
+
' "$(cat "${MEMINDEX_DATA_DIR:-$HOME/.memindex}/worker-secret")" | curl -H @- "$@"; }
|
|
29
32
|
```
|
|
30
33
|
|
|
31
34
|
This honors `MEMINDEX_WORKER_PORT` env, then `~/.memindex/settings.json`, then falls back to the per-UID default `38700 + (uid % 100)` — matching how the worker itself picks its port. Required for multi-account setups (#2101) and any user who has overridden the default port (#2103).
|
|
@@ -58,7 +61,7 @@ If a worktree is detected, use `$parent_project` (the basename of the parent rep
|
|
|
58
61
|
Use Bash to fetch the complete timeline from the memindex worker API:
|
|
59
62
|
|
|
60
63
|
```bash
|
|
61
|
-
|
|
64
|
+
memcurl -s "http://localhost:${WORKER_PORT}/api/context/inject?project=PROJECT_NAME&full=true"
|
|
62
65
|
```
|
|
63
66
|
|
|
64
67
|
This returns the entire compressed timeline -- every observation, session boundary, and summary across the project's full history. The response is pre-formatted markdown optimized for LLM consumption.
|
|
@@ -68,7 +71,7 @@ This returns the entire compressed timeline -- every observation, session bounda
|
|
|
68
71
|
- Medium project (1,000-10,000 observations): ~50-300K tokens
|
|
69
72
|
- Large project (10,000-35,000 observations): ~300-750K tokens
|
|
70
73
|
|
|
71
|
-
If the response is empty or returns an error, the worker may not be running or the project name may be wrong. Try `
|
|
74
|
+
If the response is empty or returns an error, the worker may not be running or the project name may be wrong. Try `memcurl -s "http://localhost:${WORKER_PORT}/api/search?query=*&limit=1"` to verify the worker is healthy.
|
|
72
75
|
|
|
73
76
|
### Step 3: Estimate Token Count
|
|
74
77
|
|
|
@@ -195,15 +198,15 @@ Tell the user:
|
|
|
195
198
|
|
|
196
199
|
## Error Handling
|
|
197
200
|
|
|
198
|
-
- **Empty timeline:** "No observations found for project 'X'. Check the project name with: `
|
|
201
|
+
- **Empty timeline:** "No observations found for project 'X'. Check the project name with: `memcurl -s \"http://localhost:${WORKER_PORT}/api/search?query=*&limit=1\"`"
|
|
199
202
|
- **Worker not running:** "The memindex worker is not responding on port ${WORKER_PORT}. Start it with your usual method or check `ps aux | grep worker-service`."
|
|
200
|
-
- **Timeline too large:** For projects with 50,000+ observations, the timeline may exceed context limits. Suggest using date range filtering: `
|
|
203
|
+
- **Timeline too large:** For projects with 50,000+ observations, the timeline may exceed context limits. Suggest using date range filtering: `memcurl -s "http://localhost:${WORKER_PORT}/api/context/inject?project=X&full=true"` -- the current endpoint returns all observations; for extremely large projects, the user may want to analyze in time-windowed segments.
|
|
201
204
|
|
|
202
205
|
## Example
|
|
203
206
|
|
|
204
207
|
User: "Write a journey report for the tokyo project"
|
|
205
208
|
|
|
206
|
-
1. Fetch: `
|
|
209
|
+
1. Fetch: `memcurl -s "http://localhost:${WORKER_PORT}/api/context/inject?project=tokyo&full=true"`
|
|
207
210
|
2. Estimate: "Timeline fetched: ~34,722 observations, estimated ~718K tokens. Proceed?"
|
|
208
211
|
3. User confirms
|
|
209
212
|
4. Deploy analysis agent with full timeline
|
|
@@ -32,6 +32,9 @@ If the user wants a single sweeping report, use `timeline-report` instead. This
|
|
|
32
32
|
|
|
33
33
|
```bash
|
|
34
34
|
WORKER_PORT="${MEMINDEX_WORKER_PORT:-$(node -e "const fs=require('fs'),p=require('path'),os=require('os');const uid=(typeof process.getuid==='function'?process.getuid():77);const fallback=String(38700+(uid%100));try{const s=JSON.parse(fs.readFileSync(p.join(os.homedir(),'.memindex','settings.json'),'utf-8'));process.stdout.write(String(s.MEMINDEX_WORKER_PORT||fallback));}catch{process.stdout.write(fallback);}" 2>/dev/null)}"
|
|
35
|
+
# memcurl sends the install secret without putting it in the command line
|
|
36
|
+
memcurl() { printf 'X-Memindex-Secret: %s
|
|
37
|
+
' "$(cat "${MEMINDEX_DATA_DIR:-$HOME/.memindex}/worker-secret")" | curl -H @- "$@"; }
|
|
35
38
|
```
|
|
36
39
|
|
|
37
40
|
## Workflow
|
|
@@ -55,7 +58,7 @@ echo "$parent_project"
|
|
|
55
58
|
|
|
56
59
|
```bash
|
|
57
60
|
mkdir -p .scratch
|
|
58
|
-
|
|
61
|
+
memcurl -s "http://localhost:${WORKER_PORT}/api/context/inject?project=PROJECT_NAME&full=true" \
|
|
59
62
|
> .scratch/cm-timeline.md
|
|
60
63
|
wc -l .scratch/cm-timeline.md
|
|
61
64
|
```
|
|
@@ -229,7 +232,7 @@ These rules emerged from running the pipeline end-to-end. Encode them every time
|
|
|
229
232
|
|
|
230
233
|
## Error Handling
|
|
231
234
|
|
|
232
|
-
- **Empty timeline**: project name wrong, or worker not running. `
|
|
235
|
+
- **Empty timeline**: project name wrong, or worker not running. `memcurl -s "http://localhost:${WORKER_PORT}/api/search?query=*&limit=1"` to verify.
|
|
233
236
|
- **Worker not running**: start it via your usual method or check `ps aux | grep worker-service`.
|
|
234
237
|
- **Subagent returns malformed carry-forward**: extract the carry-forward block by regex (` ```carry-forward ... ``` `) and pass forward verbatim. If missing, ask the agent to retry with the explicit instruction "your reply MUST include the carry-forward block fenced as ```carry-forward ... ``` at the very end."
|
|
235
238
|
- **One agent fails mid-pipeline**: retry that week with the same carry-forward. Don't skip — the chain breaks.
|