flash-loan-exploit-investigator-agent

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or data exfiltration logic were detected in the skill instructions. The content is purely educational and investigative.
  • [NO_CODE]: The skill consists entirely of markdown documentation and does not contain any executable scripts, binary files, or dependency manifests.
  • [PROMPT_INJECTION]: The skill contains strong safety guardrails that explicitly instruct the agent not to assist with illegal activities, live attacks, or circumventing security controls.
  • [INDIRECT_PROMPT_INJECTION]: The skill structure involves the processing of untrusted user input such as transaction hashes and external links. 1. Ingestion points: User-supplied transaction hashes, Solana signatures, and public write-up links in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: None; the skill does not define any automated tool usage or command execution capabilities. 4. Sanitization: Absent. The risk is assessed as safe because the skill is purely descriptive and does not provide an automated execution environment for the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 10:59 PM