AgentOps / Harness Evaluation

Harness Regression Evaluator

Evaluate proposed harness changes against benchmark prompts and red-team scenarios before adoption.

DocumentationAgent instructionsRequirements & limits

Use when proposed harness changes need benchmark and red-team evaluation before adoption.

Inputs

  • proposed harness change summary
  • benchmark prompts
  • red-team prompts
  • pass/fail criteria

Outputs

  • Evaluation summary
  • Regression findings
  • Adopt or reject recommendation

Before you begin

  • Reading and copying the procedure require no terminal setup. Check requirements for any packaged programs separately.

Try this first

Read the procedure and supply its documented inputs. Use the full text in an agent app or chat that supports pasted instructions, following that host's guidance.

Limits for this step
  • Instruction package only; it does not provide or authorize an executable runtime capability.
  • Safe for harness evaluation; require human approval before applying accepted harness changes.
Read full instructionsHide full instructionsSKILL.md · v0.1.0

Harness Regression Evaluator

When to use

  • Use before accepting a harness skill or policy change.
  • Use when a proposed update claims to improve safety or reliability.
  • Use to compare baseline and candidate harness behavior.

Inputs required

  • proposed harness change summary
  • benchmark prompts
  • red-team prompts
  • pass/fail criteria

Workflow

  1. Load the baseline expectations and proposed change.
  2. Run or simulate the benchmark and red-team prompts.
  3. Compare outcomes against pass/fail criteria.
  4. Approve, reject, or require more evidence.

Output format

  • Evaluation Summary
  • Passed Scenarios
  • Failed Scenarios
  • Regression Risks
  • Recommendation

Guardrails

  • Reject changes that weaken safety or widen scope silently.
  • Keep acceptance criteria explicit and deterministic.
  • Separate benchmark wins from red-team regressions.
  • Require human review if the evaluator cannot reproduce results.

Failure modes

  • If benchmark inputs are missing, do not approve the change.
  • If results are mixed, prefer rejection or more evidence.
  • If the proposal touches protected paths, require manual review.
View file metadata
---
name: harness-regression-evaluator
description: Test proposed harness updates against saved prompts and
  red-team cases to reject regressions before adoption.
---

Summary sources · Catalogue metadata, skill.yaml, SKILL.md, examples/input.json, examples/output.json · v0.1.0 · Release package

Install instruction files

These commands copy guidance into the selected runtime directories. Confirm that your agent app discovers it. Any packaged programs need separate setup and verification.

Install (Codex)

./bin/skills-hub install agentops/[email protected] --runtime codex

Install (Claude)

./bin/skills-hub install agentops/[email protected] --runtime claude

Install (Generic)

./bin/skills-hub install agentops/[email protected] --runtime generic --target ./my-agent/skills

How You Can Use This

Documentation

Resource form: Agent instructions

Read or copy the guidance. Any packaged programs have separate execution requirements and evidence.

First run: Read SKILL.md and follow the documented workflow.

Current limits

  • Instruction package only; it does not provide or authorize an executable runtime capability.
What installation and verification mean
File installation
Places package files and generated configuration in the selected directories.
Native discovery
Your agent app finds and accepts the instructions, tools or plugin registrations.
Mock execution
Exercises sample fixtures. Connected data and provider effects need a separate check.
Connected operation
Requires current configuration, provider authentication, scoped permissions and a successful operation in your intended environment.

Declared compatibility

Compatibility unknown: versions and capabilities have not been declared.

Classification: declared

Technical details and release files

Operational Summary

Execution mode: may-run-local-verification

Approval boundary: Safe for harness evaluation; require human approval before applying accepted harness changes.

Review and lifecycle

Review maturity: Experimental

Security review declaration: Not declared

Lifecycle: Active

These release declarations are separate from evidence that a workflow operates in your environment.

Runtime & Dependencies

ID: agentops/harness-regression-evaluator

Tags: evaluation, benchmarks, harness, regression

Runtimes: codex, claude, generic

Tool dependencies: 0

API dependencies: 0

Dependencies

Tools: Not documented

APIs: Not documented

Outputs

  • Evaluation summary
  • Regression findings
  • Adopt or reject recommendation