The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface Analysis:
Ingestion points: The skill is designed to ingest and process user-provided source code in both Analysis and Execution modes.
Boundary markers: Absent. The instructions do not define delimiters or provide guidance to ignore instructions that might be embedded within the code comments or logic being processed.
Capability inventory: High. In Execution Mode, the agent is instructed to 'directly make changes' to code, representing a file-write capability.
Sanitization: Absent. There are no instructions for validating or sanitizing the input code before it is processed or modified.
[NO_CODE]: The skill consists entirely of natural language instructions and metadata in SKILL.md. It does not include any executable scripts, binaries, or configuration files.
[SAFE]: The primary instructions follow standard software engineering best practices for code readability and maintainability.
[SAFE]: No network access, data exfiltration patterns, credential exposure, or obfuscated content were detected.

code-simplifier