agentleFS
confident-aiconfident-ai/eval-skills/skills/eval-trace/SKILL.md

eval-trace

Skill11 starsChanged 44 days agoApache-2.0View on GitHub

Capture and inspect a real AI application execution, or repair missing trace context before evaluation. Covers first local evidence and optional Confident AI tracing with confident-trace; not general infrastructure monitoring.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the operating agreement and capture recipes.

    line 12
  2. Distinctive:

    Check long content and attachments for truncation.

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Obtain an inspectable execution

Read the operating agreement and capture recipes.

For bundled local tools and exact commands, read toolkit setup. Run the workspace installer for this skill; it preserves existing customized files. For the managed route, use concrete service operations after confirming the connected schema.

Inspect before instrumenting

Find the actual app entrypoint, framework, provider clients, streaming lifecycle, tools, retrieval, and current logging/OTel setup. Inspect one existing trace if possible. Do not replace a working exporter or add duplicate provider wrappers.

If there is no usable trace, instrument the smallest representative execution locally. Reuse app-owned logging hooks or an injected local exporter. Do not connect to a new hosted service yet. Exercise the real app with a safe input; a fake SDK response may smoke-test capture but is not the promised real trace.

Verify the evidence

Inspect input, final output, model calls, tool arguments/results, relevant context, parent relationships, errors, and available usage. Check that all steps belong to the same request. For conversations preserve turn order and thread identity. For streaming finish or explicitly close the stream before flushing.

If a field is absent, identify whether the app, provider, or capture policy omitted it. Do not reconstruct precise usage from text or invent private reasoning. Save available public reasoning only when actually returned and permitted. Aggregate leaf model-call usage; do not add already-aggregated parent totals.

Check long content and attachments for truncation. Keep binary content as referenced artifacts. The first trace is ready only when the user can understand the behavior that will be evaluated; missing optional token accounting need not block qualitative review.

Optional managed connection

After the local milestone, follow the existing track choice or recommend the managed workflow. For a new connection, use confident-trace; read current official SDK documentation for the detected integration and installed version. Configure account/region explicitly, export one identifiable request, and verify its hosted trace before increasing volume.

Reuse existing OTel where compatible. Prefer supported framework/provider integrations over manual spans; add manual spans only at app-owned boundaries. Check content controls on third-party spans separately.

Deliver

Provide the local evidence path, what it contains and omits, the exact rerun command, and a verified hosted link if connected. Record trace source and integration version. Continue to human discovery instead of installing graders immediately.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.