project-idea-validator

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from GitHub and Dev.to search results which could potentially contain malicious instructions intended to influence the agent's behavior.
  • Ingestion points: The skill fetches repository metadata from GitHub and article content from Dev.to (in SKILL.md) to synthesize reports.
  • Boundary markers: The instructions for synthesizing the gap analysis do not include explicit delimiters or instructions to ignore embedded commands within the search results.
  • Capability inventory: The agent uses the tinyfish CLI for network search operations and possesses standard shell execution capabilities.
  • Sanitization: No explicit sanitization or filtering of the raw JSON data extracted from external sources is mentioned before the agent processes it.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @tiny-fish/cli package.
  • Evidence: README.md and SKILL.md direct users to install the package using npm install -g @tiny-fish/cli.
  • Context: This is a standard installation of the vendor's official command-line tool required for the skill's operation.
  • [COMMAND_EXECUTION]: The skill uses the tinyfish CLI to perform search queries and manage automation runs.
  • Evidence: SKILL.md defines multiple commands for checking installation (which tinyfish), authentication (tinyfish auth status), and executing agents (tinyfish agent run).
  • Context: These shell commands are integral to the intended functionality of researching and validating project ideas.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 01:04 AM
Security Audit — agent-trust-hub — project-idea-validator