Acmesh
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The shell scripts execute standard Unix utilities for file and directory management.
- Evidence:
scripts/script.shperformsmkdir,cat, andgrepoperations on a local data directory located at~/.local/share/acmesh/. - [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection through its data persistence mechanism.
- Ingestion points: User-provided strings are captured by the
addcommand inscripts/script.shand stored in a local file nameddata.log. This content is later retrieved and presented to the agent via thelist,search, andexportcommands. - Boundary markers: Absent. The skill does not utilize delimiters or instructional warnings when outputting the contents of its local data store.
- Capability inventory: The skill can read and write to the local file system and execute shell commands as defined in
scripts/script.shandscripts/acmesh.sh. - Sanitization: Absent. The skill appends raw input to the storage file without validation, escaping, or filtering.
Audit Metadata