Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

gs-ai/ACE-T

Repository files navigation

ACE-T SPECTRUM v3

ACE-T SPECTRUM v3

Python 3.11 Platform License Release

ACE-T SPECTRUM v3 is the active release of the Advanced Cyber-Enabled Threat Intelligence platform. It delivers unified ingestion, deterministic scoring, and interactive graph visualization for operational threat intelligence workflows.

Directory Map (Current and Correct)

  • scripts/: all Python source code and runtime modules (agents, src, core, db, graph, adapters)
  • graph/: viewer assets (ace_t_spectrum_3d.html, index.html, Three.js vendor files) and generated graph outputs
  • data/: generated ingest/cache directories (scaffolded with .gitkeep)
  • config/: source/feed configuration (ingest_sources.yaml)
  • db/: SQL schema artifacts (schema.sql)
  • docs/: documentation (this file, runbook, ThreatFox summary)
  • src/: environment spec (environment.yml)
  • run_graph_viewer.sh: primary launcher
  • requirements.txt / requirements.lock.txt: dependency manifests
  • outside_data/ (local-only): key files and local caches used at runtime

What v3 Provides

  • Unified graph build path for batch, streaming, and live polling modes.
  • Multi-source ingestion for incidents, infrastructure intel, and enrichment.
  • Deterministic scoring outputs (prediction_score, prediction_label).
  • Graph metadata summary (prediction_summary) for quick posture review.
  • Viewer-ready JSON artifact generation with bounded edge expansion for performance.

Release Highlights (v3 vs v2)

  • Unified internal pipeline (_build_graph_payload, _record_to_node, _predict_record) across build modes.
  • ThreatFox integrated as a first-class feed in the main graph pipeline.
  • Source controls centralized in config/ingest_sources.yaml.
  • Edge cap via ACE_T_MAX_EDGES_PER_NODE (default 24) to improve render scalability.
  • Cleaner launch surfaces with shell wrappers and script-based module execution.

Quick Start

Run from SPECTRUMv3:

conda env create -f src/environment.yml
conda activate ace-t-env
pip install -r requirements.txt

Launch viewer:

bash run_graph_viewer.sh

Run tiered ingest:

bash scripts/run_tiered_ingest.sh

Run agents:

bash scripts/run_agents.sh

Operating Modes

Examples:

python3 scripts/graph/build_graph.py
python3 scripts/graph/build_graph.py --streaming
ACE_T_LIVE_POLL_INTERVAL=300 python3 scripts/graph/build_graph.py --live

Data Source Coverage

Primary incident feed

  • ransomware.live

Infrastructure intelligence

  • abuse.ch threatfox
  • abuse.ch urlhaus
  • abuse.ch feodotracker
  • c2intelfeeds (verified + 30d)
  • montysecurity c2 tracker
  • carbon black c2

Reputation enrichment

  • blocklist_de
  • ipsum levels (3-8)

Background context

  • cisa_kev
  • optional nvd_cve (disabled by default)

Config path: config/ingest_sources.yaml
Legend/source colors: graph/data/sources.json

Output Artifacts

  • graph/graph_3d.json
  • graph/graph_3d_render.json
  • graph/data/sources.json

Environment Controls

  • RANSOMWARE_LIVE_API_KEY: override key file usage
  • ACE_T_ENABLE_STREAMING=1: enable streaming mode
  • ACE_T_FORCE_BUILD=1: force rebuild
  • ACE_T_SKIP_BUILD=1: reuse existing artifacts
  • ACE_T_LIVE_POLL_INTERVAL=<seconds>: live polling interval
  • ACE_T_MAX_EDGES_PER_NODE=<int>: edge cap (default 24)

Security Notes

  • Keep secrets in environment variables or ignored local files only.
  • Do not commit API keys, tokens, or private datasets.
  • Rotate and revoke exposed credentials immediately.

Additional Documentation

  • SPECTRUM_RUNBOOK.md: reproducible launch and operating runbook
  • THREATFOX_INTEGRATION_SUMMARY.md: ThreatFox architecture and integration details

About

Conditioned Aversion OSINT platform enabling real-time multi-source collection, correlation, and intelligence graph synthesis.

Topics

Resources

Stars

Watchers

Forks

Contributors

AltStyle によって変換されたページ (->オリジナル) /