The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external client communication data (emails, Slack messages, and quotes) to inform the tone and content of testimonials. This represents an indirect prompt injection surface.
Ingestion points: SKILL.md (Phase 1) and SKILL-OC.md (Draft-on-Behalf Rules) instruct the agent to analyze client voice data.
Boundary markers: Absent. The instructions do not define specific delimiters for separating user-provided data from agent instructions.
Capability inventory: None. The skill does not utilize tools, network operations, file system writes, or command execution.
Sanitization: Present. The 'Phase 5: Self-Critique Pass' and 'Draft-on-Behalf Rules' mandate that the agent verify the draft against the client's actual claims and style, serving as a logical filter.
[SAFE]: No malicious patterns were detected. The skill contains no code, shell commands, or obfuscated content. The external link to brianrwagner.com is a legitimate resource belonging to the author.

testimonial-collector