agent-safety

0.8.0

Idiomatic, zero-dependency, provider-agnostic Python safety layer for AI agents: least-privilege permissions, prompt/input/output guards, sandbox + SSRF guards, quotas/rate/deadline/concurrency/risk budgets, loop detection, human-in-the-loop approval, constitutional rules, action previews, honeytokens, audit/tracing, and Claude/OpenAI/Gemini/MCP tool-call integration.

License Sources Match

MIT confirmed by 2 independent sources — Python registry metadata and the LICENSE file in the package source — as of June 23, 2026.

SourceLicenseClass
Licensie (detected)
MITPermissive
PyPI (reported)
MITPermissive
Loading dependencies…
License File
Added Removed Expected
Versions
1 version
VersionLicensePublishedStatus
0.8.0 Latest ViewingMITJun 23, 2026 Scanned