The Agent Skills Directory

[SAFE]: The skill implements a read-only context retrieval mechanism. It uses the search_memories tool to fetch relevant history without performing any modifications or deletions.
[SAFE]: No evidence of data exfiltration, prompt injection, or malicious command execution was detected. The use of identifiers like user_id and app_id in search filters is standard for memory management and does not involve credential theft.
[SAFE]: The skill's behavior matches its stated purpose of priming the agent's context for new tasks. It does not engage in any suspicious network activity or use external dependencies.

context-loader