add-serper

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from the Serper API (an external untrusted source), which creates a potential vector for indirect prompt injection. Ingestion points: Search results, including snippets, titles, and news content, are fetched from the Serper API at google.serper.dev as described in SKILL.md (Step 4 and Step 7). Boundary markers: The instructions do not define the use of delimiters or 'ignore embedded instructions' warnings for isolating search results from agent instructions during the search-augmented generation flow. Capability inventory: The skill context includes tools such as write_file and run_code, which could be targeted if the agent follows malicious instructions embedded in search result snippets. Sanitization: There is no requirement for sanitizing, escaping, or validating the content retrieved from the Serper API before it is processed or summarized by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 12:30 PM
Security Audit — agent-trust-hub — add-serper