AgentClash

Feature

Agent scorecards that justify a release decision

Scorecards turn long agent runs into a reviewable verdict. AgentClash aggregates correctness, cost, latency, tool strategy, and validator evidence so baselines and candidates are easy to compare.

live eval
gate: pass

Candidate

92correct patch, low cost

Baseline

88stable reference run

Control

73missed edge case

replay timeline

1loaded task inputs and tool policy
2ran sandbox actions and captured artifacts
3scored trajectory and validator evidence
4attached scorecard and release verdict

ci verdict

Candidate clears release gate

Correctness improved, latency within budget, and required artifacts were preserved for review.

agentclash run create --follow

What scorecards include

Built for reviewable agent decisions

A scorecard should be more than a single number — it should explain why a run passed or failed.

OpenTelemetry-compatible trace import

Pinned datasets and golden test cases

Baseline versus candidate regression checks

Replay trails for tool calls, outputs, and artifacts

Scorecards for correctness, cost, latency, and evidence

CI gates for prompt, model, RAG, and tool changes

Workflow

From run to scorecard

Import the evidence

Start from OpenTelemetry traces, curated datasets, support transcripts, or a real failure your team already saw.

Pin the baseline

Record the current accepted behavior so every prompt, model, RAG, or tool change has a fair comparison point.

Replay the evidence

Inspect tool calls, outputs, artifacts, latency, cost, and judge evidence when a candidate gets worse.

Gate the release

Compare candidate and baseline runs, then fail CI before a regression reaches users.

FAQ

Agent scorecard FAQ

What appears on an AgentClash scorecard?

Correctness signals, validator evidence, cost, latency, tool usage summaries, and dimension scores your challenge pack defines.

Can scorecards compare two runs?

Yes. Baseline versus candidate comparisons are first-class for regression testing and CI gates.

Can scorecards be exported or shared?

Yes. Runs keep scorecard views in the product and support shareable evidence for reviewers.