The Agent Skills Directory

[COMMAND_EXECUTION] (LOW): The skill executes a local script ./scripts/log-skill.sh. While part of the intended workflow, executing file-system scripts is a capability that requires the script itself to be trusted.
[PROMPT_INJECTION] (LOW): Surface for Indirect Prompt Injection. The skill reads content from docs/solutions/ using grep. 1. Ingestion points: Document files in docs/solutions/ processed by grep. 2. Boundary markers: Absent; the agent receives raw grep output without delimiters or instructions to ignore content. 3. Capability inventory: Shell commands (grep, ls) and local script execution (./scripts/log-skill.sh). 4. Sanitization: None; the skill does not specify how to escape or validate instructions found within documented solutions.

compound-docs