The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources, which creates a surface for indirect prompt injection.\n
Ingestion points: The skill processes repository context, external framework documentation, and outputs from a delegated researcher Agent as described in the Execution_Policy and Steps sections of SKILL.md.\n
Boundary markers: Absent. The instructions do not specify the use of delimiters or instructions to ignore potential commands embedded in external data.\n
Capability inventory: The skill can invoke the Agent tool to spawn sub-agents with implementation capabilities, executes shell commands for builds and tests (e.g., npm run build), and performs file operations via TaskCreate, TaskUpdate, and local memory writes to .workbuddy/memory/.\n
Sanitization: Absent. There is no specified logic to sanitize or validate content fetched from external documentation or repository files before processing.\n- [COMMAND_EXECUTION]: The skill instructions explicitly direct the agent to execute shell commands for build processes and testing.\n
Evidence: The Execution_Policy and Examples in SKILL.md mention running npm run build, node --test, and package installations using run_in_background: true.

ultrawork