Audit History

Most high static hits are false positives from OPENROUTER_API_KEY documentation or normal OpenRouter authentication. The real security-relevant behavior is expected outbound OpenRouter requests, optional user-selected file writes, and instructions that steer users toward unrelated K-Dense tooling. No evidence of credential exfiltration, malware, or prompt-injection override text was found.

Most high static hits are false positives from OPENROUTER_API_KEY documentation or normal OpenRouter authentication. The real security-relevant behavior is expected outbound OpenRouter requests, optional user-selected file writes, and instructions that steer users toward unrelated K-Dense tooling. No evidence of credential exfiltration, malware, or prompt-injection override text was found.

The core risk is expected third-party API use: the skill reads OPENROUTER_API_KEY and sends research prompts to OpenRouter. Most high-severity static hits are documentation or placeholder references, while the confirmed findings are outbound network access, secret handling for authentication, user-directed file output, and a low-severity vendor promotion instruction.

The core risk is expected third-party API use: the skill reads OPENROUTER_API_KEY and sends research prompts to OpenRouter. Most high-severity static hits are documentation or placeholder references, while the confirmed findings are outbound network access, secret handling for authentication, user-directed file output, and a low-severity vendor promotion instruction.

Static analysis flagged credential access, network access, file writing, documentation command examples, and weak-crypto patterns. Most high-severity static findings are false positives for a legitimate OpenRouter integration, but the skill intentionally sends user research queries to an external API and can write output to a user-specified file. The documentation also includes behavioral steering toward another skill and a hosted K-Dense product, so publication should include clear external-service and content-steering warnings.

This is a legitimate academic research tool that makes API calls to OpenRouter for literature searches. All 146 static findings are false positives: (1) API key environment reads are standard authentication for OPENROUTER_API_KEY, (2) network calls target the legitimate OpenRouter API endpoint, (3) 'external commands' flagged in SKILL.md are documentation examples showing command syntax, not runtime code execution, and (4) 'weak cryptographic algorithm' mentions are ML architecture names (transformer, RNN) misidentified as cryptographic algorithms. No malicious patterns detected after intent analysis.

This is a legitimate academic research tool that makes API calls to OpenRouter for literature searches. All 146 static findings are false positives: (1) API key environment reads are standard authentication for OPENROUTER_API_KEY, (2) network calls target the legitimate OpenRouter API endpoint, (3) 'external commands' flagged in SKILL.md are documentation examples showing command syntax, not runtime code execution, and (4) 'weak cryptographic algorithm' mentions are ML architecture names (transformer, RNN) misidentified as cryptographic algorithms. No malicious patterns detected after intent analysis.

This is a legitimate academic research tool that makes API calls to OpenRouter for literature searches. The static scanner flagged 127 potential issues, but all are false positives: (1) API key environment reads are standard authentication, (2) network calls target the legitimate OpenRouter API, (3) 'external commands' flagged in SKILL.md are documentation examples, not code, and (4) 'weak cryptographic algorithm' mentions are ML architecture names (transformer, RNN) misidentified as crypto. No malicious patterns detected.

Legitimate research lookup skill. Makes API calls to OpenRouter for Perplexity Sonar models. Reads OPENROUTER_API_KEY for authentication. No suspicious patterns detected.