KavachBench MadB0i · COMPLETED
KavachBench: research harness benchmarking Kavach, a Rust zero-trust policy-enforcement runtime for AI-agent tool calls, against IssueTrojanBench's real prompt-injection corpus (42 attack actions, 4 categories) with static and 10 live baseline-vs-defended runs.
github.com/MadB0i/KavachBench · ★ 1 · Forks 0 · Size 5.8 MB
SUMMARY
Technologies 9
Scored 9
Observed 9
Practices 6
Evidence 9
Skips 0
COVERAGE
Analyzed 364 files · 49 commits · 0 API calls
TECHNOLOGIES & DEPTH
TypeScript LANGUAGE Depth 70
5 files · PRODUCTION
PowerShell LANGUAGE Depth 70
5 files · PRODUCTION
JSON LANGUAGE Depth 70
106 files · PRODUCTION
TOML LANGUAGE Depth 70
1 files · PRODUCTION
Markdown LANGUAGE Depth 70
47 files · PRODUCTION
YAML LANGUAGE Depth 70
26 files · PRODUCTION
Python LANGUAGE Depth 70
122 files · PRODUCTION
pip BUILD_TOOL Depth 80
1 files · CONFIGURATION
Requests LIBRARY Depth 74
22 files · PRODUCTION
PRACTICES
documentation · observedautomated_tests · observedcontinuous_integration · absentcontainerization · absentlinting · absentformatting · absent
ACTIVITY & OWNERSHIP
First commit 2026-09-07
Last commit 2026-09-30
Active months 1
Commits 49