solo-memory-audit
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script located at
${CLAUDE_PLUGIN_ROOT}/scripts/memory_map.pyusinguvorpython. This command is used to perform the core auditing logic and scan project directories. - [DATA_EXPOSURE]: The skill accesses the agent's internal configuration and memory files to calculate context consumption and identify optimization targets.
- Evidence: Accesses paths such as
~/.claude/CLAUDE.md,~/.claude/rules/, and~/.claude/projects/*/memory/MEMORY.md. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project configuration files (
CLAUDE.md,.claude/rules/*.md) which could theoretically contain malicious instructions. However, the skill's capability is limited to auditing metrics and generating a report, and the risk is considered low within the context of its primary purpose. - Ingestion points: Reads local project files and user rules (SKILL.md).
- Boundary markers: None explicitly mentioned in the instructions.
- Capability inventory: Uses
Bashto run the internal audit script. - Sanitization: Processes files for specific metrics like line counts and frontmatter glob patterns.
Audit Metadata