kstack-drift-review
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core function involves reading and analyzing documentation files, which creates a surface for indirect prompt injection if those files contain malicious instructions intended to influence agent behavior. \n
- Ingestion points: SKILL.md instructs the agent to read all instances of AGENTS.md, CLAUDE.md, README.md, and all files within the docs/ directory.\n
- Boundary markers: Absent. The instructions do not specify the use of delimiters or clear separation between external data and the agent's instruction context.\n
- Capability inventory: SKILL.md outlines the use of shell commands including git diff, git ls-files, and grep, along with repository-wide file system read access.\n
- Sanitization: Absent. There are no steps provided to sanitize or validate the documentation content before the agent extracts and processes claims for review.
Audit Metadata