The safeguards map / 3.8 Record-keeping
OpenTelemetry
Drafted with AI
Drafted with AI from the cited sources under the direction of UnGovr staff. UnGovr holds editorial responsibility for this page. How this site is made
APIs and SDKs for traces, metrics and logs, with instrumentation packages that record generative AI client and framework activity.
Where it sits
3.8 Record-keeping (primary): Records the activity of generative AI client libraries and frameworks as spans, metrics and logs, which any backend of the operator's choice can store and read.
OpenTelemetry instrumentation packages for Generative AI client libraries and frameworks in Python.
captured
emit spans, metrics, and logs according to
captured
Semantic Conventions for Generative AI (GenAI), including spans, metrics, and events for GenAI clients, MCP (Model Context Protocol), and provider-specific conventions (OpenAI, etc.).
captured
Instrument your code once using OpenTelemetry APIs and SDKs.
captured
Facts
- Concern
- securitysafety
- Names the law
- no
- Subject
- agentAI system
- Used by
- those who make itthose who run it
- Runs
- at runtime
Handoffs
produces Runtime records: OpenTelemetry spans, metrics and logs (GenAI semantic conventions)
Collector's note
The GenAI semantic conventions are a specification and an artifact; this entry is the project's software that produces the records. The repository named is the GenAI instrumentation package set.
Every quote is verbatim from a stored capture of the page linked under it. Placement is our reading of the product's own documentation; no publisher has reviewed or endorsed it. Rows CC BY 4.0. Names, marks and logos belong to their owners; logos identify each product and are not part of the dataset.