This release adds 3 notable features for engineering teams evaluating rollout.
Published 4mo
Productivity & Wikis
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
corpus-generator
deterministic-ai
enterprise-data
enterprise-datasets
synthetic-dataset-generation
Summary
AI summaryExpanded string truncation limits to 80 characters and added LLM-driven sentiment drift with crew‑AI rewriting.
Full changelog
[v1.1.1] — 2026-03-19
Changed
- String Truncation Limits (
src/,eval/): Standardized and expanded summary truncation limits from 40–60 characters to 80 characters across JIRA titles, Slack interactions, incident root causes, and PR titles to prevent critical context loss in logs and RAG retrieval. - RAG Embedding Logic (
src/memory.py): EnhancedOllamaEmbedderto support asymmetric retrieval. The system now prepends specific instruction prefixes forsearch_queryandsearch_documentto improve embedding quality for models like Stella and MXBAI. - Causal Threading Fixes (
eval/eval_harness.py): Refined_design_doc_threadsto ensure design documents are only included in evaluation chains if they possess a validcausal_chainfact, preventing broken threads in the evaluation harness. - Codebase Formatting: Applied consistent linting and multi-line dictionary wrapping across
eval/eval_harness.pyandsrc/insider_threat.pyto match project style guidelines.
Added
- LLM-Driven Sentiment Drift (
src/insider_threat.py): Replaced static text templating with a CrewAI-powered rewriting task. Disgruntled or malicious actors now use aworker_llmto authentically rewrite Slack messages, with negativity intensity scaling based on the days since the threat "onset." - Enhanced Vector Search Filtering (
src/memory.py):- Added
type_excludesupport torecall(), allowing the RAG pipeline to explicitly ignore specific artifact types (e.g., hidingpersona_skillfrom general queries). - Implemented a "causal floor" using the
sinceparameter to allow bounded timestamp filtering ($gte and $lte) within MongoDB vector searches.
- Added
- Comprehensive Memory Testing (
tests/test_memory.py): Introduced a massive expansion of the test suite (20+ new tests) covering:- Ollama instruction prefix validation.
- Mutually exclusive filter guards in
recall(). - Upsert logic for artifacts to prevent vector index duplication.
- Event-type skip lists to reduce embedding noise for high-volume, low-signal events like standard Slack messages.
Full Changelog: https://github.com/aeriesec/orgforge/compare/v0.5.0...v1.1.1
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 Orgforge
All releases →Related context
Related tools
Beta — feedback welcome: [email protected]