loc-report
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script located at
scripts/loc/loc_history.pyusingpython3. It also utilizes thersvg-convertsystem utility to transform SVG assets into PNG images. These operations involve direct shell command execution to process repository data and generate documentation artifacts. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes Swift source code from the project environment to calculate line counts. This data ingestion creates a potential attack surface for indirect prompt injection.
- Ingestion points: Swift source files throughout the codebase, which are read by the
loc_history.pyscript (SKILL.md). - Boundary markers: None mentioned; there are no explicit instructions to the agent to ignore embedded commands within the source files being analyzed.
- Capability inventory: Execution of Python scripts, shell command execution via
rsvg-convert, and file write operations toscripts/loc/loc_history.csv,scripts/loc/loc_history.svg, anddocumentation/assets/loc_history.png(SKILL.md). - Sanitization: None mentioned; the security of the processing depends on the unprovided implementation of the local Python script.
Audit Metadata