The Agent Skills Directory

[SAFE]: The skill instructions and metadata do not contain any malicious patterns, obfuscation, or unauthorized access attempts. The functionality is consistent with the stated purpose of managing persistent memory via the Honcho service.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data by instructing the agent to query a persistent memory store using tools like search and chat. While retrieved context could theoretically contain instructions intended to influence the agent, the skill provides no dangerous capabilities that could be exploited.
Ingestion points: SKILL.md (retrieval via search, get_context, and chat MCP tools)
Boundary markers: Absent
Capability inventory: None (the skill consists entirely of instructional text and does not include scripts or executable commands)
Sanitization: Absent

honcho-memory