The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override safety filters, bypass constraints, or extract system prompts were detected in the instructions.
[DATA_EXFILTRATION]: There are no network requests, credential access patterns, or sensitive file path references (e.g., .ssh, .aws) found in the skill content.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to the ingestion of external data files.
Ingestion points: The skill instructions specifically search for and read .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md to establish context (SKILL.md).
Boundary markers: There are no boundary markers or instructions to treat the content of these files as untrusted data.
Capability inventory: The skill does not currently define or use tools with high-risk capabilities such as arbitrary command execution, network exfiltration, or file modification.
Sanitization: No sanitization or validation of the content read from the context files is implemented.

onboarding