weekly-composer-dependency-audit
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands (
composer outdated,composer audit) to perform its primary task. These commands are restricted to read-only auditing operations and do not include package installation or system modification. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing and summarizing untrusted data from project configuration files.
- Ingestion points: The skill reads contents from
composer.jsonandcomposer.lockin the local project directory (SKILL.md). - Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potential instructions embedded within the ingested metadata.
- Capability inventory: The agent has the capability to execute shell commands and generate natural language reports based on the findings.
- Sanitization: No evidence of metadata validation, filtering, or escaping is present before the data is processed by the agent.
Audit Metadata