The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to manage a local directory structure for the generated reports.
Evidence: mkdir -p findings/ and file creation instructions in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied data regarding security vulnerabilities, which presents a potential injection surface.
Ingestion points: User input provided via the $finding-report command interface.
Boundary markers: The skill employs structured markdown tables and fenced code blocks to delimit user-provided content.
Capability inventory: Capabilities are restricted to local file system operations (directory creation and markdown file writing).
Sanitization: The instructions explicitly mandate slugifying titles into kebab-case for use as filenames, which serves as a security control against path traversal attacks.

finding-report