memory-kit
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes dynamic context injection (exclamation-backtick syntax) in
SKILL.mdto execute shell commands such aswcanddateto display the status of theMEMORY.mdfile upon loading. Additionally, it executes a local shell scriptscripts/memory-share.shto automate git staging, committing, and pushing operations. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it is designed to read and summarize potentially untrusted content from
MEMORY.mdto restore session context. - Ingestion points:
MEMORY.mdis read by the agent during the/memory-loadroutine described inSKILL.md. - Boundary markers: Absent; the instructions do not define delimiters or warnings to prevent the agent from following instructions that might be embedded within the persisted memory.
- Capability inventory: The skill has access to
Bash(git:*),Read,Write, andEdittools as defined in theallowed-toolsfrontmatter ofSKILL.md. - Sanitization: Absent; the skill summarizes and adopts the state from the file without content validation or sanitization.
Audit Metadata