spire-studio / cloakbot
A privacy-first AI agent that sanitizes your prompts locally with a local LLM before forwarding to remote LLM APIs.
Why this job: says: local llm
What it does
Frontier LLM use is now load-bearing — but the data that crosses the wire is non-revocable. CloakBot moves enforcement before the wire: a local privacy kernel on Gemma 4 E2B that detects sensitive spans, assigns stable typed placeholders, redacts images, chunks long documents, and restores outputs locally from a per-session vault. The remote LLM is interchangeable — Claude, GPT, and Gemini all accept the sanitised stream unchanged.
Get started
pip install --pre cloakbot
# 0.2.1 beta (privacy kernel); drop --pre once 0.2.1 is stable- Maintenance
- Active
- Last commit 4 months ago
- Stars
- 205
- On GitHub
- Licence
- Custom
- Read before business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.2.0
- Released 5 months ago
- Built with
- Python
- 81% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- None
- No issues filed
- Pull requests
- 15 merged
- 0 waiting
- Releases
- 2
- 2 in the last year
More: languages, topics and project record
Languages
- Python 81%
- TypeScript 18%
- CSS <1%
- Shell <1%
Topics
- ai-security
- llm
- local-first
- privacy
- privacy-preserving-ai
- prompt-sanitization
- trustworthy-ai
Project record
- JobLocal LLMs
- StatusListed
- Commits65
- Forks0
- Watching0
- Stars in 28 days+1
- CreatedApr 10, 2026
- First seen by usNov 7, 2026
- CheckedNov 9, 2026