monologue-notes
Fail
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: Downloads and executes a shell script from a remote URL (https://raw.githubusercontent.com/EveryInc/monologue-toolkit/main/install.sh) by piping the output of curl directly into sh in SKILL.md and references/install.md.
- [COMMAND_EXECUTION]: Uses the shell to execute various commands using the monologue CLI for searching, listing, and retrieving notes.
- [INDIRECT_PROMPT_INJECTION]: Processes data from note transcripts and summaries which may contain malicious instructions.
- Ingestion points: Note transcripts, summaries, and metadata retrieved through CLI commands in SKILL.md.
- Boundary markers: Instructions to use quotation marks for verbatim content and to clearly label synthesis in SKILL.md.
- Capability inventory: Terminal command execution for API interaction via the monologue CLI.
- Sanitization: No explicit content sanitization or validation is defined for the data returned from the external API.
Recommendations
- HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/EveryInc/monologue-toolkit/main/install.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata