The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted user input through the $ARGUMENTS placeholder, which constitutes a potential surface for indirect prompt injection.
Ingestion points: External issue descriptions are ingested via the $ARGUMENTS variable within the Mission section of SKILL.md.
Boundary markers: The input is encapsulated within <issues> XML-style tags to provide structural separation from the skill's instructions.
Capability inventory: The skill has the authority to read project files, write analysis reports to the .ai/workspace/analysis/ directory, and perform file modifications to implement bug fixes.
Sanitization: There is no explicit evidence of sanitization or content filtering for the data provided in $ARGUMENTS beyond the use of delimiters.

fix-fast