paid-measurement-loop

Use when the user asks to "read back" a paid campaign change, "did this ad change work", or "compare ROAS/CPA before and after"; reads ROAS/CPA against a control over a fixed readback window and returns a Promote / Keep-testing / Rollback / Unproven readback decision with the math delegated to roi-calculator. Not for RQS scoring or veto adjudication — use ad-account-auditor; not for the ROI ratio math — use roi-calculator; not for cross-channel rollups — use performance-analyzer. 付费广告复盘/ROAS回看/投放效果归因

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

85%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable readback protocol with a clear sequenced workflow and validation checkpoints. Its only weakness is repetition of the non-auditor boundary guard across several sections, which adds tokens without adding instruction.

Suggestions

Consolidate the auditor-boundary guard: state the 'do not emit verdict/veto_count/cap/score_state/DONE-BLOCK' list once (e.g. in step 7) and reference it from the opening paragraph, Data Sources, and Next Best Skill instead of restating it each time.

The opening paragraph repeats the roi-calculator / ad-account-auditor / performance-analyzer disambiguation already present in the frontmatter description; consider trimming it to a single sentence pointing at the Skill Contract for the full boundary.

Tighten step 7 by separating the Unproven-and-handoff rule from the repair procedure into two short clauses, removing the inline restatement of which fields are auditor-only.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of general-knowledge padding, but the auditor-boundary guard ('not the auditor', 'do not emit verdict/veto_count/cap/score_state...') is restated across the opening paragraph, Data Sources, step 7, and Next Best Skill, so it could be tightened. Not a 3 because not every token earns its place; not a 1 because it assumes Claude's competence and avoids explaining known concepts.	2 / 3
Actionability	Provides concrete executable commands with real paths — 'python3 "${CLUDE_PLUGIN_ROOT}/scripts/connectors/ledger.py" record <campaign> --source paid --data ...' plus 'ledger.py diff' and 'ledger.py trend' — and explicit decision criteria; as an instruction-skill the delegated math is justified rather than missing.	3 / 3
Workflow Clarity	A numbered 8-step sequence with explicit checkpoints (step 1 learning-phase stop, step 7 signal-integrity → Unproven + repair + restart window) and a 'Done when' checklist, matching the anchor for clear sequence with validation steps and a feedback loop.	3 / 3
Progressive Disclosure	Well-organized into Quick Start, Skill Contract, Data Sources, Instructions, Reference Materials, and Next Best Skill, with one-level-deep, descriptively-labeled references (measurement-protocol.md, roas-benchmark.md, roi-calculator, etc.) keeping the body a lean overview.	3 / 3
	Total	11 / 12 Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that names concrete actions, surfaces natural trigger phrasing, and sharply distinguishes itself from three neighboring skills. It answers both what and when without padding.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'reads ROAS/CPA against a control over a fixed readback window and returns a Promote / Keep-testing / Rollback / Unproven readback decision with the math delegated to roi-calculator' — matching the anchor for several specific concrete actions.	3 / 3
Completeness	Explicit 'Use when...' trigger clause answers when, and the action clause answers what, satisfying the anchor for clearly answering both what AND when with explicit triggers.	3 / 3
Trigger Term Quality	Covers natural user phrasing — 'read back a paid campaign change', 'did this ad change work', 'compare ROAS/CPA before and after' — alongside ROAS/CPA, giving good coverage of terms users would actually say.	3 / 3
Distinctiveness Conflict Risk	Three explicit 'Not for ... use <other skill>' clauses (ad-account-auditor, roi-calculator, performance-analyzer) carve a clear niche unlikely to trigger the wrong skill.	3 / 3
	Total	12 / 12 Passed

Validation

75%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 14 suspicious	Warning
referenced_paths_exist	Referenced path issues: 1 missing, 1 deeper-than-1-level	Warning

	Total	12 / 16 Passed

Repository: aaron-he-zhu/aaron-marketing-skills
Path: ad/scale/paid-measurement-loop/SKILL.md
Commit: 4a77c48

Reviewed: 9 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.