alterlab-yok-tez

Warn

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions direct the user to install a tool from a specific third-party GitHub repository (github.com/saidsurucu/yoktez-mcp) which is not an officially verified or trusted organization.
  • [REMOTE_CODE_EXECUTION]: The skill recommends using uvx to install and run code directly from a remote Git repository (uvx --from git+https://github.com/saidsurucu/yoktez-mcp yoktez-mcp). This pattern executes unvetted code from the internet at runtime.
  • [DATA_EXFILTRATION]: The skill references a hosted connector service at https://yoktezmcp.fastmcp.app/mcp. Utilizing this service involves sending search parameters and potentially receiving data through an external, third-party domain.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it retrieves and processes untrusted data from an external repository including thesis titles, abstracts, and metadata.
  • Ingestion points: Data returned by the search_yok_tez_detailed and get_yok_tez_document_markdown tools from the YÖK repository (as described in SKILL.md and references/endpoints.md).
  • Boundary markers: None identified; the skill does not instruct the agent to use specific delimiters or to ignore instructions embedded within the retrieved thesis content.
  • Capability inventory: The skill has access to powerful tools including Bash, WebFetch, Write, and Edit (defined in SKILL.md).
  • Sanitization: There is no evidence of content sanitization or validation for the retrieved data before it is presented to or processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 15, 2026, 03:28 PM
Security Audit — agent-trust-hub — alterlab-yok-tez