The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied metrics and reads existing files (such as a 'prior board deck') to generate new reports. This data ingestion creates a surface where malicious instructions within the input files could potentially influence the agent's behavior. \n- Ingestion points: Workflow step 4 in SKILL.md directs the agent to read prior board decks provided by the user. \n- Boundary markers: There are no instructions to use delimiters or ignore instructions found within the ingested data. \n- Capability inventory: The skill is configured to use Read, Write, Glob, Grep, and Bash tools within its execution environment. \n- Sanitization: The instructions do not include steps for sanitizing or filtering input from user-provided files before interpolation into the output.

board-deck-generator