User repository on GitHub
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Demos & Experiments
Jev Semantic Graph Explorer — Threat Intelligence
A Python research project that uses Jev (TypeSafe AI, System 1) with Neo4j to traverse cybersecurity knowledge graphs probabilistically, choosing promising graph edges and scoring nodes against a goal instead of generating text.
- Stars
- 0
- Language
- Python
- Reading time
- 2 min
- Published by
- Community
- Author
- pauloribeiro16
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open sourceRequires API key
Highlights
- Uses Noul as a boolean stop criterion for whether the current node satisfies the intended semantic goal.
- Choice deterministically picks which adjacent Neo4j edge is the most promising next hop.
- Score supplies an ordinal goal-alignment heuristic h(n) for A* or best-first search.
- The knowledge graph integrates CWE 4.19 with 969 weaknesses and MITRE ATT&CK with 836 techniques.
- Targets sub-50ms structured judgments, contrasting them with slower, hallucination-prone LLM text.
Watch out
README reviewed 2026-10-17 in Portuguese; the captured text stops in the CAPEC section, so Neo4j setup, Jev prompts, evaluation results and licence are not visible. Metadata shows 0 stars and no licence.
Questions about this entry
Short answers drawn from the linked project or page.
A Python research project for probabilistic traversal of cybersecurity knowledge graphs, using Jev with Neo4j to decide which graph nodes and edges to explore next.
Noul acts as a boolean stop criterion, evaluating the probability that the current node satisfies the intended semantic goal before traversal continues.
Choice performs categorical selection among the adjacent Neo4j edges, deterministically picking the branch deemed most promising for the next hop.
Score supplies an ordinal goal-alignment heuristic h(n), intended to guide informed search algorithms such as A* or best-first search.
It integrates CWE 4.19 with 969 software weaknesses, MITRE Enterprise ATT&CK with 836 techniques and sub-techniques, and CAPEC attack patterns.
The README argues Jev performs structured, probabilistic micro-judgments in under 50ms, avoiding the slowness, graph-node hallucinations and token cost of System 2 LLM text generation.
The Portuguese README stops in the CAPEC section, so Neo4j setup, Jev prompts, traversal results, evaluation and licensing are not available; metadata shows 0 stars and no licence.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
YouTube: Kubernetes admission control with a decision model, fail-closed.
Hacker News: SafeDep Gryph plus Jev checks for compromised coding-agent behavior.
2 pointsDetecting Compromised AI Coding Agents with Jev and GryphRead the thread on Hacker NewsX: Jev and Laya on security queries
X: Testing Jev for prompt injection
More Jev resources like this
Two Jev demos over SprintPulse QA data: one scores release risk with a 10-level rubric evaluated against a reference score, the other diagnoses the main risk cause with a 9-label choice question, both sending the same per-ticket signals to Jev's System One API.
Jev Demos & Experiments#community#github#python
Community
avi350751/jev-qa-analyzerPython
avi350751/jev-qa-analyzerPythonA small SeleniumBase project that sends a page's reduced HTML to Jev and gets back three structured answers: whether a CAPTCHA is present, a 0-100 confidence score and which of six types it is. The README warns scores vary between runs and are not calibrated probabilities.
Jev Demos & Experiments#community#github#captcha
Community
mdmintz/jev-testingPython
mdmintz/jev-testingPythonA Python web app that sends every support ticket to Jev (typesafe/jev-1.13) and to configured comparison models through OpenRouter, showing team routing, urgency, severity, sentiment and draft replies side by side with latency and cost.
Jev Demos & Experiments#community#github#python
Community
nikojim/jevPython
nikojim/jevPythonA Python first-project that turns messy free-text smartphone reviews into Flipkart-style per-topic star ratings, powered by Jev via the TypeSafe SDK: 14 questions per review — mention detection plus satisfaction — across 7 topics such as camera, battery and display.
Jev Demos & Experiments#community#github#python
An open-source browser 3D character split between two models: GPT-Live-1 talks with you over WebRTC while Jev reads the live transcript and picks Annie's face, gestures and body language in about 120 ms, with code gates on every answer.
Jev Demos & Experiments#community#github#python
Community
rehan-remade/open-anniePython
rehan-remade/open-anniePythonA Chinese-language e-commerce customer-service demo on FastAPI, LangGraph and React: adaptive retrieval routing with hybrid BM25-plus-vector search, and intent classification switchable between TypeSafe Jev's structured choice interface and DeepSeek.
Jev Demos & Experiments#community#github#python
Community
gbh1234/Adaptive-ragPython
gbh1234/Adaptive-ragPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.