wikidata-knowledge
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses curl to interact with Wikidata's REST and SPARQL endpoints and python3 to parse the resulting JSON. These operations are essential for the skill's primary function of information retrieval.
- [REMOTE_CODE_EXECUTION]: Output from the Wikidata API is piped to a python3 process for parsing. This is evaluated as a safe data-processing pattern because the python script is statically defined within the skill and the data source is a well-known, trusted organization.
- [PROMPT_INJECTION]: The 'Data Integrity Rule' section instructs the agent to rely strictly on API responses rather than its own training data. This is a reliability instruction aimed at reducing hallucinations.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection via data returned from Wikidata.
- Ingestion points: Untrusted content from Wikidata enters the agent's context through curl results.
- Boundary markers: No specific delimiters are used to isolate the data.
- Capability inventory: The agent possesses the capability to run shell commands (curl, python3).
- Sanitization: There is no evidence of filtering or sanitization of the retrieved content before it is processed by the agent.
Audit Metadata