This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+8 more
ReleasePort's take
Moderate signalv1.2.0 introduces a feedback vault collection with tools and REST endpoints, optional Teams adaptive‑card notifications after feedback notes, daily activity logs mirroring key events, folder‑based node typing, and configurable search tuning via new environment variables.
Why it matters: The release adds observable telemetry (feedback vault, daily logs) and operational controls (Teams alerts, searchable cache tuning), enabling developers and SREs to monitor and react to system behavior more effectively.
Summary
AI summaryDefault search mode switched to hybrid; QMD container crash and empty‑result caching bugs fixed.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Breaking | High |
Changes default search mode to hybrid from keyword. Changes default search mode to hybrid from keyword. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Feature | Low |
Adds feedback vault collection with tools and REST endpoints. Adds feedback vault collection with tools and REST endpoints. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Feature | Low |
Adds optional Teams notifications posting adaptive cards after feedback notes. Adds optional Teams notifications posting adaptive cards after feedback notes. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Feature | Low |
Adds daily activity log mirroring vault_write, vault_ingest, and compile events. Adds daily activity log mirroring vault_write, vault_ingest, and compile events. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Feature | Low |
Adds folder‑based node typing for agents/, sessions/, and daily/ without suffixes. Adds folder‑based node typing for agents/, sessions/, and daily/ without suffixes. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Feature | Low |
Adds configurable search tuning via CORTEX_QMD_CACHE_TTL_SECONDS and CORTEX_QMD_SEARCH_TIMEOUT_SECONDS. Adds configurable search tuning via CORTEX_QMD_CACHE_TTL_SECONDS and CORTEX_QMD_SEARCH_TIMEOUT_SECONDS. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Performance | Medium |
Memoizes QMD path index per graph mutation to avoid rebuilding on every search. Memoizes QMD path index per graph mutation to avoid rebuilding on every search. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Performance | Low |
Offloads vault_read note parsing to a separate thread for non‑blocking concurrent MCP requests. Offloads vault_read note parsing to a separate thread for non‑blocking concurrent MCP requests. Source: granite4.1:30b@2026-06-13-audit Confidence: low |
— |
| Bugfix | Medium |
Fixes QMD container failure due to missing qmd binary after install. Fixes QMD container failure due to missing qmd binary after install. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Bugfix | Medium |
Fixes hybrid searches being silently truncated by hardcoded 30s timeout. Fixes hybrid searches being silently truncated by hardcoded 30s timeout. Source: llm_adapter@2026-06-13 Confidence: high |
— |
| Bugfix | Medium |
Fixes cached search failures by no longer caching empty result sets. Fixes cached search failures by no longer caching empty result sets. Source: llm_adapter@2026-06-13 Confidence: high |
— |
Full changelog
Added
- Feedback vault collection: New
feedback/folder withvault_feedbackandvault_list_feedbacksMCP tools plus REST endpoints, capturing user/agent feedback on vault quality withstatusandrelated_paths. - Microsoft Teams notifications: Optional
TEAMS_WEBHOOK_URLposts an adaptive card after each new feedback note; optionalTEAMS_APP_BASE_URLadds a "View in Cortex" link. - Daily activity log: Every
vault_write,vault_ingest, and successful compile is mirrored intodaily/<UTC-date>.mdas a greppable## [HH:MM] event | summaryentry with a[[wiki-stem]]wikilink. Writes todaily/,feedback/, and.cortex/are excluded to avoid self-reference. - Folder-based node typing: Files under
agents/,sessions/, anddaily/are classified asAGENT_DEF,SESSION, and the newNodeType.DAILYwithout requiring a filename suffix. Frontmattertype:and the existing.agent.md/.session.mdsuffixes still take precedence/work. - Configurable search tuning:
CORTEX_QMD_CACHE_TTL_SECONDS(result-cache TTL, default 30s) andCORTEX_QMD_SEARCH_TIMEOUT_SECONDS(search request timeout, default 120s).
Changed
- Default search mode is now
hybrid(waskeyword). Clients that don't pass an explicitmodenow get the best-quality BM25 + vector + re-ranking results. SetCORTEX_QMD_SEARCH_MODE=keywordto restore the previous fast-but-shallow default. Latency is mitigated by the result cache and the HTTP QMD container.
Fixed
- QMD container returned no results: The image installed
qmdvianpm install -g ., which created a global symlink into a build directory deleted in the same layer — leaving a danglingqmdbinary so every search failed withspawn qmd ENOENTand silently returned[]. Now installed from a packed tarball with a build-timeqmd --versionsmoke check. - Hybrid searches silently truncated: QMD HTTP searches hit a hardcoded 30s timeout on CPU-only hosts and returned
[]while QMD was still working. Timeout is now configurable (default 120s). - Cached search failures: Empty result sets (transport errors/timeouts) are no longer cached, preventing a transient failure from blanking search for the full cache TTL.
Performance
- Memoized QMD path index:
build_path_index_from_graphis cached per graph mutation and no longer rebuilt on every search/read. - Non-blocking reads:
vault_readoffloads note parsing to a thread so concurrent MCP requests no longer serialize behind file I/O.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Cortex
All releases →Related context
Beta — feedback welcome: [email protected]