The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted content from external architecture and requirements files without adequate protection.\n
Ingestion points: The architecture_file and requirements_file provided as inputs are read into the agent's context (SKILL.md).\n
Boundary markers: The skill does not use specific delimiters or instructions to treat the ingested file content as untrusted or to ignore embedded instructions.\n
Capability inventory: The skill can execute a local Python script to perform file reads and generates a detailed analytical report based on the input.\n
Sanitization: There is no evidence of sanitization or validation of the input file content before it is processed by the agent.\n- [COMMAND_EXECUTION]: The workflow involves executing a local Python script (.claude/skills/bmad-commands/scripts/read_file.py) to read input files. This creates a dependency on the local environment's scripts.\n- [DATA_EXFILTRATION]: The skill allows the agent to read any file path provided via the architecture_file parameter. This capability could be used to expose sensitive system or configuration files if the agent is directed to an unintended path.

architecture-review