tenten-link-building
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests article titles and descriptions from the vendor's database via the search_content tool. While processing external text is an injection surface, the risk is mitigated by the data originating from a private, authenticated source. 1. Ingestion points: Data enters the context through search results in SKILL.md. 2. Boundary markers: No explicit delimiters are used for API output. 3. Capability inventory: The agent is instructed to modify article text to insert links. 4. Sanitization: The skill relies on manual relevance checks by the agent.
- [DATA_EXFILTRATION]: The skill facilitates network communication with
index.tenten.devfor searching and content submission. These operations are core to the skill's functionality and target the official infrastructure of the skill author (tentenco).
Audit Metadata