vultr/openclaw
1
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-05-06 18:30:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(memory): skip qmd vectors in lexical mode

Browse Source
This commit is contained in:
Peter Steinberger
2026-04-27 14:09:32 +01:00
parent 6a0dc3a9bc
commit b181930c23
6 changed files with 94 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/cli/memory.md
View File

@@ -51,7 +51,7 @@ openclaw memory index --agent main --verbose
`memory status`:
- `--deep`: probe vector + embedding availability. Plain `memory status` stays fast and does not run a live embedding ping.
- `--deep`: probe vector + embedding availability. Plain `memory status` stays fast and does not run a live embedding ping. QMD lexical `searchMode: "search"` skips semantic vector probes and embedding maintenance even with `--deep`.
- `--index`: run a reindex if the store is dirty (implies `--deep`).
- `--fix`: repair stale recall locks and normalize promotion metadata.
- `--json`: print JSON output.

13
docs/concepts/memory-qmd.md
View File

@@ -51,13 +51,15 @@ present.
## How the sidecar works
- OpenClaw creates collections from your workspace memory files and any
configured `memory.qmd.paths`, then runs `qmd update` + `qmd embed` on boot
and periodically (default every 5 minutes).
configured `memory.qmd.paths`, then runs `qmd update` on boot and
periodically (default every 5 minutes). Semantic modes also run `qmd embed`.
- The default workspace collection tracks `MEMORY.md` plus the `memory/`
tree. Lowercase `memory.md` is not indexed as a root memory file.
- Boot refresh runs in the background so chat startup is not blocked.
- Searches use the configured `searchMode` (default: `search`; also supports
`vsearch` and `query`). If a mode fails, OpenClaw retries with `qmd query`.
`vsearch` and `query`). `search` is BM25-only, so OpenClaw skips semantic
vector readiness probes and embedding maintenance in that mode. If a mode
fails, OpenClaw retries with `qmd query`.
- If QMD fails entirely, OpenClaw falls back to the builtin SQLite engine.
<Info>
@@ -164,6 +166,11 @@ runs as a service, create a symlink:
**First search very slow?** QMD downloads GGUF models on first use. Pre-warm
with `qmd query "test"` using the same XDG dirs OpenClaw uses.
**BM25-only QMD still trying to build llama.cpp?** Set
`memory.qmd.searchMode = "search"`. OpenClaw treats that mode as lexical-only,
does not run QMD vector status probes or embedding maintenance, and leaves
semantic readiness checks to `vsearch` or `query` setups.
**Search times out?** Increase `memory.qmd.limits.timeoutMs` (default: 4000ms).
Set to `120000` for slower hardware.

2
docs/reference/memory-config.md
View File

@@ -449,6 +449,8 @@ Set `memory.backend = "qmd"` to enable. All QMD settings live under `memory.qmd`
| `sessions.retentionDays` | `number` | -- | Transcript retention |
| `sessions.exportDir` | `string` | -- | Export directory |
`searchMode: "search"` is lexical/BM25-only. OpenClaw does not run semantic vector readiness probes or QMD embedding maintenance for that mode, including during `memory status --deep`; `vsearch` and `query` continue to require QMD vector readiness and embeddings.
OpenClaw prefers the current QMD collection and MCP query shapes, but keeps older QMD releases working by falling back to legacy `--mask` collection flags and older MCP tool names when needed.
<Note>