The Agent Skills Directory

[SAFE]: The skill is a text-processing utility intended to summarize discussion history and decision logs. It does not perform network operations, execute code, or modify system configurations.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by reading from docs/discussions/ and conversation history. While this represents a potential vector for processing untrusted data, the skill's capabilities are limited to standard file-reading tools (Read, Grep, Glob), and it lacks high-privilege operations (like shell execution or network exfiltration) that could be exploited via injection. Explicit labeling for assumptions like (my assumption, not discussed) acts as a basic boundary marker for AI-generated content.

decision-summary