start
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
bashtool to runwhichcommands for identifying installed security tools likesemgrep,bandit, andgitleaksin the local PATH.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by reading and processing untrusted data from project files.\n - Ingestion points: The agent reads various manifest files (e.g.,
package.json,go.mod) and source code files (e.g.,Dockerfile,serverless.yml).\n - Boundary markers: The instructions do not define specific delimiters or guidelines for the agent to ignore potentially malicious instructions embedded within the codebase being scanned.\n
- Capability inventory: The skill employs
Glob,Grep,Read, andBash(used forwhichcommands). It also writes assessment results to the local file system (.appsec/start-assessment.json).\n - Sanitization: There is no evidence of content sanitization or validation for the data ingested from the project files.
Audit Metadata