qe-coverage-analysis
Warn
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The 'Run History' section in SKILL.md instructs the agent to execute a dynamic JavaScript snippet using 'node -e' to update a local 'run-history.json' file. This template uses unquoted placeholders (STATEMENTS, BRANCHES, GAPS) that, if populated with unsanitized data from coverage tools, could allow for local command or script injection within the node environment.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and analyze external test coverage reports (Istanbul, c8, lcov), which presents a surface for indirect prompt injection.
- Ingestion points: Coverage report files in the local environment are read and processed by the agent in SKILL.md.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potentially malicious content embedded within these reports.
- Capability inventory: The skill possesses the capability to execute shell commands ('aqe'), perform dynamic execution ('node -e'), and conduct file system operations.
- Sanitization: The instructions do not define any sanitization or validation steps for the content of the coverage reports before they are analyzed or logged to the history file.- [COMMAND_EXECUTION]: The skill relies on a local CLI tool ('aqe') and shell commands for coverage analysis, gap detection, and reporting, which requires the agent to execute arbitrary commands in the local shell environment.
Audit Metadata