Remove ~170MB of model files from the repository (LFS + committed).
Models are now downloaded at runtime from Hugging Face on first use
and cached to the OS cache directory with progress reporting and retries.
- Add internal/models/download.go: runtime downloader with retry, progress bar, checksums
- Remove go:embed for ONNX models (keep tokenizers embedded)
- Use file-based ONNX session loading instead of byte-slice
- Add scripts/download-models.sh for dev/CI model setup
- Update Makefile with setup-models target
- Update workflow-prepare.sh to download models in CI
- Set lfs: false in all CI workflows
- SHA256: bge=828e14..., cross-encoder=5d3e70...
MCP server (5 fixes):
- Move semaphore acquisition inside goroutine so main loop stays
responsive when all slots are taken
- Add 10s write timeout to sendResponse to prevent pipe deadlock
when Claude Code pauses reading stdout
- Send fallback JSON-RPC error when json.Marshal fails instead of
silently swallowing the error and leaving caller waiting forever
- Silence unknown notification methods (req.ID == nil) instead of
sending unsolicited error responses that may desync the host
- Return MCP isError content for tool failures instead of top-level
JSON-RPC error, matching the MCP specification
Vector/embedding (3 fixes):
- Move EmbedBatchWithContext call before writeMu.Lock in AddDocuments
so ONNX inference runs outside the write lock
- Replace singleflight.Do with DoChan + ctx select in both
getOrComputeEmbedding and UnifiedSearch so callers can bail out
independently when their context expires
- Add activeQueries atomic counter; skip cache warming when user
queries are in-flight; reduce warming timeout from 5s to 2s
Hooks (4 fixes):
- Cap EnsureWorkerRunning to 15s hard deadline with context; reduce
StartupTimeout from 30s to 10s; reduce port-in-use retries
- Fix nil dereference panic in user-prompt hook when initResult is
nil (non-JSON worker response); use comma-ok assertions
- Use package-level hookClient/healthClient with DisableKeepAlives
to prevent FD leaks in short-lived hook processes
- Set SysProcAttr{Setpgid: true} to detach worker from hook process
group, preventing kill-cascade from Claude Code
Worker/DB (3 fixes):
- Replace os.Exit(0) in MCP config watcher with context cancellation
for clean protocol shutdown
- Add 60s context.WithTimeout around ProcessObservation calls in
processAllSessions to prevent hung CLI subprocesses from blocking
the queue processor forever
- Set explicit PRAGMA wal_autocheckpoint=1000 and add PASSIVE WAL
checkpoint to Optimize() to prevent checkpoint stalls
Adds 20+ regression tests across all fix areas.
Root cause: plugin registered as directory source in known_marketplaces.json,
which gets wiped on CLI updates. Now registers in extraKnownMarketplaces
(settings.json) as a GitHub source — same mechanism caveman/context-mode use.
Binaries install to ~/.claude-mnemonic/bin/ instead of the Claude-managed
plugins directory. Thin wrapper scripts in the repo let the marketplace
clone find them. Nothing gets cleaned up when Claude refreshes its cache.
Also fixed along the way:
- ONNX Runtime 1.24.3 → 1.26.0 (API v25 mismatch broke all embedding tests)
- Vector client leaked on DB reinit, processQueue had a race on sessionManager
- reloadConfig called os.Exit(0) bypassing graceful shutdown
- Removed dead QueryRowWithTimeout that leaked contexts
- Added tests for graph/watcher/maintenance/update (all were at 0%)
* refactor(hooks): simplify hook execution with shared context
- [x] Extract BaseInput struct to eliminate duplicate fields across hooks
- [x] Create RunHook handler pattern for session-start and user-prompt
- [x] Create RunStatuslineHook for fast statusline rendering without worker startup
- [x] Add HookContext struct to pass port, project, CWD, SessionID to handlers
- [x] Add db/interface.go with ObservationReader/Writer interfaces
- [x] Add comprehensive conflict management tests in sqlite/conflict_test.go
- [x] Add vector client tests for Count, ModelVersion, NeedsRebuild, GetStaleVectors
- [x] Add FilterByThreshold helper tests for query result filtering
- [x] Make handlers_test more robust for network-dependent update checks
- [x] Update package versions in UI
* Move to GORM + general cleanup
* feat(mcp): add observation relations discovery and scoring integration
- [x] Add find_related_observations MCP tool for discovering related observations by confidence
- [x] Integrate scoring calculator and recalculator into MCP server initialization
- [x] Add pattern, relation, and session stores to MCP server dependencies
- [x] Register MCP server in Claude Code settings during plugin installation
- [x] Update install scripts (bash, PowerShell) to configure MCP server settings
- [x] Switch plugin manifest files to template-based versioning (plugin.json.tpl, marketplace.json.tpl)
- [x] Update all MCP server tests to pass new dependency parameters