The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests untrusted data from project specification files and test command outputs, creating a surface for indirect prompt injection.
Ingestion points: Data is read from files in the specs/ directory and from the output of test runners like npm test or pytest (as described in the verification process).
Boundary markers: The instructions do not define clear delimiters or include warnings to ignore instructions embedded within the ingested data.
Capability inventory: The skill utilizes shell command execution (test runners), file reading (cat), and file writing (touch).
Sanitization: There is no evidence of sanitization or validation of the ingested content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill requires the agent to execute various shell commands, including test runners and file system utilities (fd, cat, touch). These commands are necessary for the skill's primary function of verifying implementation quality but represent a capability to run code defined within the project repository.

verification-before-completion