teams-meeting-pipeline
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and summarize Microsoft Teams meeting transcripts and recordings, which are untrusted external data sources. This introduces a potential surface for indirect prompt injection if the ingested content contains malicious instructions intended to manipulate the agent's output.
- Ingestion points: Untrusted meeting data is ingested via
hermes teams-pipeline fetchandhermes teams-pipeline runcommands. - Boundary markers: The instructions do not specify explicit delimiters or "ignore embedded instructions" warnings for the agent when processing meeting content.
- Capability inventory: The skill utilizes the terminal tool to execute shell commands (
hermes), which perform meeting summarization, action item extraction, and delivery to Microsoft Teams channels. - Sanitization: No specific sanitization or filtering logic is described within the skill instructions; protection relies on the underlying CLI implementation and model guardrails.
Audit Metadata