project-idea-validator
Audited by Snyk on Apr 24, 2026
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The prompt includes and instructs use of a literal-looking API key (sk-tinyfish-...) in example commands and env var settings, which causes the agent to output or propagate a secret verbatim and thus creates exfiltration risk.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to run TinyFish against public GitHub and Dev.to search URLs (see "Step 1 — Search GitHub" and "Step 2 — Search Dev.to" with tinyfish agent run --url "https://github.com/search?q=..." and "...dev.to/search?q=..."), ingesting untrusted, user-generated content and using it to drive synthesis and verdicts—allowing third‑party content to materially influence the agent's actions.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill calls TinyFish at runtime to fetch and extract live content from external pages (e.g. https://github.com/search?q=&type=repositories&s=stars&o=desc and https://dev.to/search?q=), and then directly injects that fetched JSON/content into the agent's synthesis step—i.e., runtime-fetched external content is used as required input to control the agent's outputs.
HIGH W008: Secret detected in skill content (API keys, tokens, passwords).
- Secret detected (high risk: 1.00). The documentation contains a high-entropy, fully-formed API key value used verbatim in multiple example environment variables and settings: sk-tinyfish-k7KpC3TypJEAwK4j1yN0Ww7yB8h7VPLg
This is not a placeholder (not "sk-xxxx" or truncated), not an obvious low-entropy example, and matches the pattern of usable API keys. It appears directly in the docs and therefore should be treated as a real credential. Recommend removing the exposed key, rotating it immediately, and replacing examples with a placeholder like YOUR_TINYFISH_API_KEY.
Issues (4)
Insecure credential handling detected in skill instructions.
Third-party content exposure detected (indirect prompt injection risk).
Unverifiable external dependency detected (runtime URL that controls agent).
Secret detected in skill content (API keys, tokens, passwords).