The Agent Skills Directory

[SAFE]: The skill content consists of informational documentation and templates for business processes. No executable scripts, network requests, or unauthorized file access patterns were detected. Analysis of the instructions and reference documents reveals that the content is focused solely on providing strategic advice for lead management and sales alignment.- [PROMPT_INJECTION]: The skill instructions include a pattern for reading local project context files to personalize recommendations, which creates a potential surface for indirect prompt injection. 1. Ingestion points: The agent is directed to read .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md at the start of the interaction (SKILL.md). 2. Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present when reading these files. 3. Capability inventory: The skill does not limit the agent's tool access via frontmatter, meaning the agent's full capability set is available. 4. Sanitization: No sanitization or validation of the context file content is performed.

revops