Skills
skills/tinyfish-io/tinyfish-cookbook/project-idea-validator/Gen Agent Trust Hub

project-idea-validator

Fail

Audited by Gen Agent Trust Hub on Apr 24, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: A hardcoded API key sk-tinyfish-k7KpC3TypJEAwK4j1yN0Ww7yB8h7VPLg is included in the SKILL.md file within multiple examples for setting environment variables.
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands using the tinyfish command-line interface to perform web searches and data extraction.
  • [EXTERNAL_DOWNLOADS]: The skill documentation requires the global installation of the @tiny-fish/cli package from the npm registry.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from external sources.
  • Ingestion points: Untrusted data is ingested from search results on github.com and dev.to through the tinyfish tool (identified in SKILL.md).
  • Boundary markers: No explicit boundary markers or 'ignore embedded instructions' warnings are provided to separate fetched data from the agent's task instructions.
  • Capability inventory: The skill uses the tinyfish agent run command to execute browser automations and search operations (identified in SKILL.md).
  • Sanitization: There is no evidence of sanitization or validation performed on the retrieved search results before they are synthesized into the final report.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 24, 2026, 10:19 AM