compact-memory
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill accesses and modifies local files within the
.claude/rules/directory to perform hygiene tasks such as compression and pruning. These actions are confined to the local project environment and are consistent with the skill's primary function of context management.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes historical session data which could contain instructions from previous untrusted inputs.\n - Ingestion points: Reads from
.claude/rules/memory-sessions.mdand.claude/rules/memory-decisions.mdin SKILL.md.\n - Boundary markers: No delimiters or ignore instructions are specified for the processed content.\n
- Capability inventory: The skill modifies local markdown files and writes project facts to the memory graph via the
mcp__memory__create_entitiestool.\n - Sanitization: The instructions do not define sanitization or validation protocols for the ingested data.
Audit Metadata