relatr-elo-plugin-author
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONNO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown documentation, templates, and reference materials. It does not include any executable scripts or binary files.
- [COMMAND_EXECUTION]: The skill provides instructions for the user to execute CLI commands using the
@contextvm/relopackage (e.g.,relo check,relo build,relo publish). These commands include placeholders for user-supplied secrets like the--secflag for private keys. - [PROMPT_INJECTION]: The skill processes user-supplied Elo source code as part of its 'Review' and 'Debug' workflows, creating a surface for indirect prompt injection.
- Ingestion points: User-provided Elo source code in 'Review' and 'Debug' tasks.
- Boundary markers: Not present.
- Capability inventory: The agent generates Elo code, manifest metadata (Nostr kind 765), and shell commands (
relo). - Sanitization: No sanitization of the input Elo source code is performed before analysis.
Audit Metadata