The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by instructing the agent to use untrusted user input (task descriptions and IDs) to perform file system operations and generate documentation.\n
Ingestion points: User task requests and task descriptions used for folder naming in SKILL.md.\n
Boundary markers: Absent; the workflow does not define delimiters or instructions to ignore embedded commands in user-provided data.\n
Capability inventory: The skill uses mkdir -p for directory creation and npm test for verification, as seen in SKILL.md and references/tasks.md.\n
Sanitization: Absent; the instructions do not specify validation or escaping of user input before it is used in shell commands.\n- [COMMAND_EXECUTION]: The skill directs the agent to execute standard development commands, specifically mkdir -p to scaffold documentation structures and npm test to verify implementation. These actions are aligned with the skill's primary purpose but rely on input derived from user requests.\n- [NO_CODE]: The skill consists entirely of markdown documentation and contains no executable scripts or binaries.

sdd