Use Case
Kiro + Cursor Rules
Steering files tell your AI what to do on every task. Nobody is reading them for security.
CRITICAL
Kiro steering file injection — 2 confirmed attack vectors in .kiro spec definitions — AI confirmed
CHK-119
Kiro injection detected
CHK-116–118
cursor rules checkers
177k★
affected repo (everything-claude-code)
Use Case
autoApprove + steer block = silent exfiltration on every task.
CHK-119 and CHK-120 confirmed 2 attack vectors in everything-claude-code. The .kiro spec declared autoApprove and write access to **/* — then embedded an exfiltration instruction in the steer block.
CHK-119: Kiro steering file injection
CHK-120: dangerous permissions + auto-execute in .kiro/specs
Full repo walk — catches specs in any subdirectory
LLM verification on CRITICAL findings
How it works
01
Connect your GitHub org
OAuth in 30 seconds. AISS discovers every MCP server, skill file, hook, and agent config across all repos.
02
22 modules scan in parallel
CVE lookup, secret scanning, auth checking, tool description analysis, skill file parsing — all concurrent, all hand-written.
03
LLM verifies high-severity
Critical and high findings go to an LLM verifier before reporting. No false positives reach your CISO.
04
Gate, alert, or export
Block in CI via SARIF. Send to SIEM via NDJSON. Export CycloneDX SBOM. Enforce allowlist/blocklist policy.
Press coverage
VentureBeat
Anthropic Skill scanners passed every check. The malicious code rode in on a test file.
VentureBeat
No publicly documented scanner operates outside the assumption that the threat lives in SKILL.md.
CrowdStrike · RSAC 2026
ClawHavoc — 1,184 malicious skills confirmed in the wild. The attack surface is the skill layer.