The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied text for editorial review, creating a potential attack surface for indirect prompt injection.
Ingestion points: Processes arbitrary text strings provided by the user as input for editing tasks.
Boundary markers: The skill does not define explicit delimiters or boundary markers (such as XML tags or unique markers) to separate user content from the agent's core instructions.
Capability inventory: The skill is configured with Read, Write, Edit, Grep, and Glob tools in its frontmatter, allowing the agent to perform file modifications based on instructions derived from the input.
Sanitization: There is no evidence of input validation, filtering, or escaping of the user-supplied text before it is analyzed by the editorial layers.
[EXTERNAL_DOWNLOADS]: The README documentation provides installation instructions that involve external resources from the author's infrastructure.
Evidence: Manual installation steps include git clone https://github.com/iamursky/sokrati. These instructions target the author's official GitHub repository and are standard practice for skill distribution.
[NO_CODE]: The skill is comprised entirely of instructional Markdown (SKILL.md) and reference documentation (KNOWLEDGE.md). It contains no executable scripts (e.g., .py, .js, .sh) or binary files, which inherently limits its ability to perform unauthorized system operations.

abrevia