Back to skill

Security audit

sciverse academic retrieval

Security checks for vulnerabilities and agentic risk

Overview

The skill provides its advertised academic searches through Sciverse, with expected sharing of search queries.

Search queries and filters go to Sciverse, so omit confidential details unless sharing is intended. Keep the default HTTPS endpoint and protect your API token. Frontmatter formatting may need correction before loading. Remote service behavior and retention practices were not verified.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 49)May include surrounding context.

md
**Invoke**: `node scripts/semantic_search.mjs '<JSON args>'`

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 115)May include surrounding context.

md
**Invoke**: `node scripts/get_resource.mjs '<JSON args>'`

The RAG workflow sends user-derived queries to Sciverse without a privacy warning.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

The workflow instructs the agent to run semantic_search on a natural-language query using the external Sciverse service. Although the file identifies the service and its authentication requirements, it does not warn that user questions or potentially confidential research details included in queries will be transmitted to that third party.

Content

No source excerpt is available for this finding.