The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or dangerous command executions were detected.
[DATA_EXPOSURE]: The skill instructs the agent to read from .agents/product-marketing-context.md or .claude/product-marketing-context.md. This is a legitimate request for project-specific context and does not involve accessing sensitive system files or credentials.
[INDIRECT_PROMPT_INJECTION]: There is a potential attack surface where the skill ingests untrusted data from the aforementioned context files. If these files were to contain malicious instructions, the agent might follow them. However, since these files are local to the user's project, the risk is considered low.
[PROMPT_INJECTION]: No evidence of role-play bypasses, safety filter overrides, or system prompt extraction attempts was found.

launch-strategy