memclaw
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation guides the user to install the @memclaw/memclaw plugin using the command 'openclaw plugins install @memclaw/memclaw'. This involves downloading external code from a package registry, which is a standard procedure for this tool.\n- [PROMPT_INJECTION]: The skill operates as a memory retrieval system, creating a vulnerability surface for indirect prompt injection.\n- Ingestion points: External data enters the agent's context through tools such as cortex_search, cortex_recall, and cortex_get_content (references found in SKILL.md and references/tools.md).\n- Boundary markers: The instructions lack clear delimiters or warnings to the agent to ignore any potential instructions embedded within the retrieved memory content.\n- Capability inventory: The skill allows the agent to read from and write to a local memory structure, including the ability to browse the filesystem via cortex_ls and commit session data via cortex_commit_session.\n- Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from memory before it is presented to the agent.
Audit Metadata