unstuck
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform file system operations, including creating a directory structure at
${MAKERSKILLS_CONFIG:-$HOME/.config/makerskills}/unstuck/archive/and writing/appending markdown files to that location to archive roadblock data. - [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface where it processes user-provided descriptions of 'goals,' 'walls,' and 'attempts' to generate archival filenames and content. This creates a surface where malicious user input could potentially influence filesystem operations, though the risk is localized to the specific configuration directory.
- Ingestion points: Step 1 (The goal, The wall, Attempts, Source of the no) and Step 7 (slug generation from the wall description).
- Boundary markers: None present in the provided instructions to delimit user content from instructions.
- Capability inventory: File system write access and directory creation in the user's home directory (Step 7).
- Sanitization: The instructions do not explicitly mandate sanitization of user-provided strings before they are used as filenames (slugs).
Audit Metadata