The Agent Skills Directory

[PROMPT_INJECTION]: The skill acts as an intermediary that processes untrusted data (user requirements) to produce executable plans, creating a surface for indirect prompt injection.
Ingestion points: User requirements and specifications provided as input to the skill (processed in SKILL.md).
Boundary markers: Absent; the skill does not define delimiters or specific instructions to the agent to disregard embedded instructions within user-provided specs.
Capability inventory: The generated plans (defined in SKILL.md) contain Python code snippets and shell commands for pytest and git commit intended for execution.
Sanitization: Absent; the skill does not specify any validation, escaping, or sanitization logic for content interpolated from user specs into the generated plan files.
[COMMAND_EXECUTION]: The skill templates and encourages the preparation of shell commands (e.g., pytest, git add, git commit) for execution. While the skill itself does not invoke a shell, it is designed to facilitate command execution through its multi-agent workflow.

writing-plans