The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external configuration files, which creates a surface for indirect prompt injection attacks where malicious instructions could be embedded in data processed by the agent.
Ingestion points: The skill uses Read, Grep, and Glob tools to ingest DAST configuration files (e.g., af-plan.yaml, zap.yaml, burp-config.json, and CI/CD workflow files) as described in Step 1 and Step 2 of SKILL.md.
Boundary markers: The skill includes a dedicated 'Prompt Injection Safety Notice' section that explicitly instructs the agent to treat configuration content as untrusted data and ignore any text that appears to be a prompt or instruction.
Capability inventory: The skill is restricted to Read, Grep, and Glob capabilities, allowing it to search and inspect local files but preventing it from performing network operations or executing arbitrary system commands.
Sanitization: While the skill provides architectural instructions to the agent to ignore embedded instructions, it does not implement automated sanitization or escaping of the file content before it is processed in the agent's context.

dast-config