jenkinsfile-validator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external and untrusted Jenkinsfiles and Groovy shared library files. If an attacker embeds malicious natural language instructions inside pipeline comments or string literals, it could potentially influence the behavior of the AI agent reviewing the linter logs or code findings.
- Ingestion points: Target validation files are passed as command-line arguments to
validate_jenkinsfile.shorvalidate_shared_library.shwhere they are read and parsed. - Boundary markers: There are no explicit prompt boundaries or defensive instructions guiding the agent to ignore natural language instructions contained within the code or comments of files being validated.
- Capability inventory: The skill executes local bash scripts that invoke text processing utilities (
grep,sed,awk,find) and makes platform tool calls (mcp__context7__resolve-library-id,mcp__context7__query-docs) to resolve plugin usage. - Sanitization: The validation scripts process file contents purely as text streams through utility patterns (e.g., filtering out comments and tracking brace depths), ensuring file contents are never executed as active shell commands. No natural language prompt sanitization is performed on the files.
Audit Metadata