Terminal
Terminal/Skill
Adds tracing to an LLM app or agent with the confident-trace SDK, so each model call, retrieval, and tool call shows as a span.
By Confident AI (DeepEval)
Works with
npx skills add confident-ai/confident-trace --skill confident-tracingUse this skill to instrument an AI application—an LLM app, agent, RAG pipeline, or chatbot—with the confident-trace SDK so its execution is visible span by span in Confident AI. Pick a supported integration when one exists, fall back to custom spans otherwise, give each span a meaningful type, and add useful trace context.
This skill stops at producing well-formed traces. Attaching evaluation metrics
and running evals is the deepeval skill's job.
Instrument only the AI parts of the system: agent loops and planning, LLM calls, retrieval/vector search, and tool calls. Do not trace non-AI software such as web servers, CRUD backends, database layers, or infrastructure. If the target has no LLM, agent, retrieval, or tool-calling component, this skill does not apply.
deepeval and confident-otel Skillsconfident-tracing)—instrument a Python or TypeScript app
with the confident-trace SDK and its integrations.deepeval skill—build pytest evaluation suites: datasets, metrics,
traced evals, deepeval test run, and iteration.confident-otel skill—export raw, vendor-neutral OpenTelemetry traces to
Confident AI without the confident-trace package.confident-trace installed.CONFIDENT_API_KEY for export to Confident AI.references/integrations.md and the exact repository integration
documentation for what was detected. Prefer automatic instrumentation.references/tracing.md.llm, retriever, tool, agent, or
custom) and capture useful inputs and outputs.thread, user, and customer objects carry extra properties such as
names, tags, and metadata.init().turn() for conversation turns that must start separate traces; a
thread ID associates turns but does not merge their OpenTelemetry traces.deepeval; raw
OpenTelemetry export belongs to confident-otel.| Topic | File |
|---|---|
| SDK instrumentation: initialization, custom spans, context, and lifecycle | references/tracing.md |
| Integration selection and supported provider/framework index | references/integrations.md |