The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and rewrite untrusted external data such as emails, landing pages, and reports. Because the execution environment grants the agent file-manipulation permissions, there is a potential for indirect prompt injection where malicious instructions embedded in the processed documents could attempt to hijack the agent's actions.
Ingestion points: User-provided professional texts for revision (SKILL.md).
Boundary markers: Absent; the instructions do not specify delimiters to separate user data from agent instructions or warn the agent to ignore embedded commands.
Capability inventory: The skill is configured with Read, Write, Edit, Grep, Glob, and AskUserQuestion tools (SKILL.md frontmatter).
Sanitization: Absent; the skill lacks mechanisms to filter or escape instructions found within the input text.
[EXTERNAL_DOWNLOADS]: The README.md file contains installation instructions directing users to git clone from the author's repository (github.com/iamursky/sokrati). As these resources are owned by the skill's author, they are considered standard vendor resources.

abbrevia