The Agent Skills Directory

[COMMAND_EXECUTION]: Executes local system commands including grep for pattern matching, and language-specific tools like tsc, mypy, go build, and cargo check for verification. These operations are standard for development workflows and consistent with the skill's primary purpose.
[DATA_EXPOSURE_AND_EXFILTRATION]: Reads local source code files to extract type definitions. The skill writes analysis results to a local report directory and does not exhibit any network exfiltration behavior or access to sensitive system configuration files.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (source code) to generate consolidation reports and modify type definitions.
Ingestion points: Reads source files with extensions .ts, .tsx, .py, .go, and .rs using grep (defined in SKILL.md).
Boundary markers: None identified in the generated report format.
Capability inventory: File reading (grep), file writing/modification (type consolidation), and command execution (build/type-check tools) as defined in SKILL.md.
Sanitization: No explicit sanitization or escaping of extracted type definitions is mentioned before report generation or consolidation.

cleanup-types