building-multi-connector-agent
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the
airbyte-agent-sdkpackage, which is a vendor-provided library for connector integration. - [INDIRECT_PROMPT_INJECTION]: The skill scaffolding creates an attack surface where malicious data from external sources could influence the agent's behavior.\n
- Ingestion points: Untrusted data from Jira and Slack enters the agent context via the
jira_executeandslack_executetools defined inSKILL.md.\n - Boundary markers: The provided example system prompts do not include delimiters or instructions to the model to treat external data as untrusted or to ignore instructions embedded within retrieved content.\n
- Capability inventory: The agent is configured with tools to read issues and post messages, providing a capability path for data manipulation or exfiltration if the prompt is subverted.\n
- Sanitization: No sanitization or filtering of the external data is performed in the provided code snippets before it is passed to the language model.
Audit Metadata