manage-memory
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to perform file system operations such as checking line counts and verifying file existence. These operations are essential for the skill's purpose of managing theMEMORY.mdfile and its associated topic documents. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes content from memory files which are populated by past, potentially untrusted, interactions.
- Ingestion points: The skill reads
MEMORY.mdand other files within the project's local memory directory (~/.claude/projects/*/memory/). - Boundary markers: Absent; the instructions do not include delimiters or specific instructions to ignore embedded commands within the memory content.
- Capability inventory: The skill is granted
Bash,Write, andEditpermissions, allowing for broad system interaction and file modification. - Sanitization: Absent; Step 6 contains a bash loop that extracts strings from markdown links using
grepand passes them directly to thelscommand without sanitization or quoting, creating a significant command injection vector if the memory file contains malicious link targets.
Audit Metadata