Agent Share of Choice Evaluator
Evaluate brand progression across retrieval, consideration, recommendation, selection, transaction, completion, and reuse in agent-mediated tasks.
DocumentationRequirements & limits
Use when a team needs a repeatable task-panel evaluation of paid, earned, and executable brand participation.
Inputs
The released example shows these fields. Check the guide for which are required.
panel_versiontaskssurfacesbrand
Outputs
- Stage evidence
- Route comparison
- Confidence and next-test report
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.
- Evaluation only; do not publish comparative claims without methodology and human review.
Read full instructionsHide full instructions
Agent Share of Choice Evaluator
When to use
Use this skill for the situations described in its frontmatter routing description.
Workflow
- Load a versioned task panel and freeze the evaluation conditions.
- Run each task consistently across the selected agent surfaces.
- Capture evidence for retrieval, understanding, consideration, recommendation, paid response, selection, transaction, completion, and reuse.
- Label observed, synthetic, platform-reported, and independently verified evidence separately.
- Calculate stage rates only where denominators and samples are explicit.
- Compare paid, earned, and executable routes without collapsing them into one rank.
- Report failures, uncertainty, and the next experiment.
Output
- task panel version and run conditions
- stage evidence by task
- stage rates with sample sizes
- route comparison
- evidence quality and confidence
- failure modes and next tests
Guardrails
- Do not claim a universal agent ranking.
- Do not treat one model, prompt, geography, or session as representative.
- Do not merge paid exposure with earned recommendation.
- Do not call platform conversion data independently verified.
View file metadata
---
name: agent-share-of-choice-evaluator
description: Evaluate whether a brand is retrieved, understood, considered, recommended, selected, transacted with, and used successfully across a stable panel of agent-mediated customer tasks. Use for AI visibility, conversational advertising, agentic commerce, or paid-versus-earned discovery experiments.
---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 marketing/[email protected] --runtime codex
Install (Claude)
./bin/skills-hub install marketing/[email protected] --runtime claude
Install (Generic)
./bin/skills-hub install marketing/[email protected] --runtime generic --target ./my-agent/skills