clawdtm-advisor
Warn
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The skill downloads content from
clawdtm.com, an unverified third-party domain not included in the trusted sources list. - COMMAND_EXECUTION (MEDIUM): The skill instructs the agent to use
curlto fetch data and then write files to the local workspace (./skills/{slug}/). - REMOTE_CODE_EXECUTION (MEDIUM): The purpose of the skill is to install executable content (other agent skills) onto the local machine. Once installed and the session is restarted, this code is executed.
- INDIRECT_PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection because it processes untrusted skill descriptions and metadata from a remote API without sanitization or boundary markers.
- Ingestion points: API responses from
clawdtm.com/api/v1/skills/searchand.../install(processed in SKILL.md). - Boundary markers: Absent; the agent is instructed to present raw results to the user.
- Capability inventory: File-writing to workspace, network requests via
curl. - Sanitization: None detected; the agent processes and displays raw strings from the API.
Audit Metadata