feed-operations

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents procedures for bulk data ingestion from external sources, creating an attack surface for indirect prompt injection if the ingested content is subsequently processed by an AI agent.
  • Ingestion points: docs/feed-clients.md includes examples of fetching and piping data from external URLs into the vespa feed command.
  • Boundary markers: The documentation does not provide specific delimiters or instructions for the agent to isolate or ignore potentially malicious instructions within the data being fed.
  • Capability inventory: The skill enables the agent to execute shell commands using the vespa CLI and curl for interacting with the database.
  • Sanitization: The documentation does not specify methods for sanitizing the ingested document data.
  • [EXTERNAL_DOWNLOADS]: The documentation contains examples illustrating how to fetch data from remote endpoints.
  • Evidence: docs/feed-clients.md demonstrates fetching data from https://example.com/data.jsonl using curl.
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands to manage documents in a Vespa cluster.
  • Evidence: SKILL.md and docs/feed-clients.md list numerous commands using the vespa CLI (e.g., vespa document put, vespa feed) and curl for standard REST operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 09:25 AM