SOCMIntelligence

Multi-platform SOCMINT / OSINT web tool.

What it is

SOCMIntelligence collects public data from many platforms, turns each result into charts, finding tables and a relationship graph, and merges everything into a single cross-platform identity graph. Because the same account gets the same node ID everywhere, an email, domain or username that shows up in two different scans is automatically linked — so you can see who is who across services.

Modules

Each module reads public data only (or an official API where a key is provided). There is no credential/session scraping of private accounts.

Features

• Relationship graph — a D3 force graph that merges every successful scan. Cross-platform matches (shared email, domain, username), pivoting (double-click a node to scan it in the right module), search, type filters, PNG/JSON export.
• Timeline — every account's posts, comments, uploads and account-creation events on one axis, plus an hour-overlap heatmap and hour-similarity table.
• Identity matching — a 0–100 “same person?” score per account pair with reasons (verified/cryptographic link, shared email, username & display-name similarity, shared personal domain, avatar pHash, activity hours). Pairs ≥ 40 are drawn on the graph.
• Comparison — put 2+ scanned accounts of the same module side by side (KPI table + bar chart, highest value highlighted).
• Analyst marks — mark graph nodes Verified / Suspicious / Dismissed, add notes, add manual nodes and links. All saved to the case.
• Case management — create a case, activate it, and scans/marks/notes are saved automatically. Import/export a .socmint package.
• Change tracking (Monitoring) — rescan targets on a schedule and record what changed (fields, new/removed links).
• Evidence integrity — every scan's raw data is hashed (SHA-256 over canonical JSON) at collection time; a verify button proves whether it was altered.
• Exports — HTML / PDF report, Excel (.xlsx), GraphML (Gephi/yEd), GEXF (Gephi), Maltego CSV, JSON.
• Bilingual UI — Turkish / English toggle (default Turkish); reports, spreadsheets, chart labels, date & number formats all follow the selected language.
• Settings page — API keys, proxy / Tor / proxy-list rotation and rate limits, all managed from the UI instead of .env.

Quick start

python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
uvicorn server:app --host 0.0.0.0 --port 8000

Then open http://localhost:8000 — API docs at http://localhost:8000/docs.

Next runs only need:

source .venv/bin/activate && uvicorn server:app --port 8000

Docker: docker compose up --build (single service).

Configuration

All configuration lives on the Settings page (saved to data/settings.json; .env is still read, Settings wins):

• API keys — stored server-side, never returned to the browser (only “set” + last 4 chars are shown).
  • YOUTUBE_API_KEY — required for the YouTube module.
  • X_BEARER_TOKEN — X (Twitter) API v2. Without it the X module shows clearly-labelled sample data so you can preview the output before adding a token.
  • STEAM_API_KEY, STACKEXCHANGE_KEY, GRAVATAR_API_KEY, GITLAB_TOKEN — optional, unlock extra data.
• Proxy / Tor — off, Tor, a single custom proxy, or a proxy list that rotates round-robin per scan (spreads rate limits).
• Rate — parallel scan count and delay between requests.

Notes

• X (Twitter): the free X plan is very limited — even with a token, post history may come back empty on some plans; the module labels that case.
• Telegram reads only the public t.me/s/ preview (no login).
• Gravatar receives only the MD5/SHA-256 hash of an email, never the email itself.
• TikTok frequently rate-limits/bot-blocks; the module retries and distinguishes a real “not found” from a temporary block.
• Time distributions are UTC. TikTok follower “growth” is an estimate (marked as such), not real history.
• backend/app/tools/youtube_osint.py ships with a hardcoded fallback API key — replace it with your own from Settings and remove that line.

License

Apache License 2.0 — see LICENSE.