The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted external specifications or requirements to generate implementation plans. This creates a surface for indirect prompt injection, where malicious instructions hidden in a project spec could influence the agent's behavior or be incorporated into the generated plan.
Ingestion points: The skill is triggered by a 'spec or requirements for a multi-step task' as described in SKILL.md.
Boundary markers: There are no explicit instructions or markers to delimit external content or warn the agent to ignore embedded instructions.
Capability inventory: The skill instructs the agent to write files to the local file system (docs/superpowers/plans/) and generate/execute commands for git and pytest.
Sanitization: No sanitization or validation of the input specification is mentioned or enforced.
[COMMAND_EXECUTION]: The skill facilitates the execution of shell commands through the generated plans.
Evidence: The SKILL.md file provides templates for tasks that include git add, git commit, and pytest commands. These are standard development tools, but they represent the operational capabilities enabled by the skill's instructions.

writing-plans