deepline-plays-review

Use this skill when a human needs to review a Deepline Play result and hand feedback, labels, or approval back to the agent for revision, evaluation, comparison, or bounded iteration. Triggers on “put this run in a Sheet,” “review these results,” “read my feedback,” “make this a standing rule,” “never regress on this case,” “compare these runs,” “keep improving,” or “add the revised run to the same spreadsheet.” Skip ordinary Play authoring or execution with no human review loop, generic spreadsheet work, and one-off CSV export.

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

88%Weight 40%Scale 1-5

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

A strong, highly actionable workflow skill with excellent sequencing and validation checkpoints. The main weakness is a lightly over-explanatory evaluation primer and a mostly monolithic structure that underuses separate reference files.

Suggestions

Trim general evaluation methodology Claude already knows (holdout-vs-development discipline, dropping-missing-cases caveat, deterministic-grader taxonomy) to the Deepline-specific packaging and scorecard details.

Consider splitting the routing tables and error-routing reference into a separate RECOVERY.md or ROUTING.md referenced one level deep, so SKILL.md reads more as an overview.

The 'Interpret feedback and revise' classification table duplicates intent already captured in the 'Route the request' table; consolidate to reduce token cost.

Dimension	Reasoning	Score
Conciseness	Largely efficient operational prose with executable blocks and no padding of basic concepts, but the evaluation section teaches general ML wisdom Claude already knows (holdout discipline, separating missing vs. incorrect cases, deterministic graders) that could be trimmed.	4 / 5
Actionability	Copy-paste-ready bash/jq blocks with env-var guards (SKILL_DIR, RUN_ID), concrete tool invocations, and a referenced presentation script cover the common cases fully.	5 / 5
Workflow Clarity	A clearly sequenced route → loop → run/review → revise → evaluate → checkpoint flow with explicit validation checkpoints ('pilot on one or two rows before scaling,' 'Never overwrite a reviewed tab,' 'Freeze the evaluation basis'), a stale-feedback re-read rule, and an error-routing table for recovery.	5 / 5
Progressive Disclosure	Well-organized into clearly headed sections with one one-level-deep, clearly signaled bundle reference (scripts/review-sheet-presentation.mjs, verified present), but the body is largely a single inline monolith with limited use of separate reference files.	4 / 5
	Total	18 / 20 Passed

Description

100%Weight 40%Scale 1-5

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

An exemplary description: concrete capabilities, natural trigger phrases, explicit what-and-when framing, and a skip clause that cleanly separates it from adjacent skills. It matches the rubric's top anchors on every dimension.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'review a Deepline Play result and hand feedback, labels, or approval back to the agent for revision, evaluation, comparison, or bounded iteration' — giving comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	Explicitly answers what ('review a Deepline Play result and hand feedback, labels, or approval back to the agent...') and when (a 'Triggers on...' clause with concrete phrases), hitting the 5-anchor exactly.	5 / 5
Trigger Term Quality	Eight natural user phrases are quoted verbatim ('put this run in a Sheet,' 'review these results,' 'read my feedback,' 'make this a standing rule,' 'never regress on this case,' 'compare these runs,' 'keep improving,' 'add the revised run to the same spreadsheet'), matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	A clear niche (human review of Deepline Play results) with a 'Skip...' clause that excludes ordinary Play authoring, generic spreadsheet work, and one-off CSV export, minimizing overlap with sibling skills.	5 / 5
	Total	20 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: getaero-io/gtm-eng-skills
Path: skills/deepline-plays-review/SKILL.md
Commit: 3bb8968

Reviewed: 3 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.