extract-source-sample

Given the path to a finished content-goose ad-run folder, extract everything that defines that ad — recipe shot list, VO script, characters, voices, world, atom-skills, master mp4 — and emit a `source-sample.json` in the exact shape the `upload-ad-sample` skill writes to the Goose Ads library. Also links every character and voice to the central character library at `<repo-root>/assets/character-library/` (repo-root derived from the run-dir, not a hardcoded path), and if a character isn't in the library yet, adds it first then links. Use when the user wants to remix one of their existing ads — this skill produces the source JSON that the script-rewriting step and `remix-ad` consume.

Quality

81%

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

77%

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

The body is highly actionable and sequenced with strong validation checkpoints for a fragile batch/read-only operation. Its main weaknesses are duplicated schema content between the inline body and the unbundled reference file, and a missed link to that reference.

Suggestions

Replace the inline source-sample.json schema (lines 104–151) with a one-line pointer to references/source-sample-schema.md, e.g. 'See [source-sample-schema.md](references/source-sample-schema.md) for the exact output shape and invariants' — removes the duplication and signals the reference.

Trim justifying prose that isn't instruction (e.g. 'an agent adapts, a script would brittle out') to tighten conciseness without losing the agent-executed rationale.

Add a couple of natural trigger phrasings ('reuse this ad', 'swap the brand on this ad', 'prep an ad for remix') alongside 'remix' to broaden trigger-term coverage.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and assumes Claude's competence on content-goose internals, but it inlines the full source-sample.json schema (lines 104–151) that is duplicated verbatim in references/source-sample-schema.md, and includes justifying prose ('an agent adapts, a script would brittle out') — matching the 'mostly efficient but could be tightened' anchor rather than fully lean.	2 / 3
Actionability	Concrete file paths to read, exact JSON shapes, executable commands ('git lfs fetch --include=...', 'git lfs checkout', 'file <path>'), filename→role mapping tables, and explicit matching rules make this instruction-only skill copy-paste-ready despite having no code, matching the fully-actionable anchor.	3 / 3
Workflow Clarity	A clear 4-step sequence (read the run → build JSON → link library → write output) with explicit validation checkpoints (materialize LFS before reading binaries, the per-PNG 'file' variant audit, refuse-to-fabricate guardrails, a summary that surfaces null links and missing/LFS assets) and error-recovery feedback loops in the failure-modes section, matching the explicit-checkpoints anchor.	3 / 3
Progressive Disclosure	Sections are well-organized, but the full output schema is inlined in the body while an equivalent references/source-sample-schema.md exists yet is never linked or signaled from the SKILL — content that should live one level deep is inline, matching the 'references present but not clearly signaled; content that should be separate is inline' anchor.	2 / 3
	Total	10 / 12 Passed

Description

85%

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

The description is concrete, complete, and clearly scoped to a distinct niche with both a 'what' and an explicit 'when' clause. Its only weakness is trigger-term breadth, leaning almost entirely on 'remix' with little coverage of synonymous phrasings.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'extract everything that defines that ad — recipe shot list, VO script, characters, voices, world, atom-skills, master mp4', 'emit a source-sample.json', 'links every character and voice to the central character library', 'adds it first then links' — matching the multi-action anchor.	3 / 3
Completeness	Explicitly answers both 'what' (extract + emit source-sample.json + link characters/voices to the library) and 'when' ('Use when the user wants to remix one of their existing ads'), matching the both-answered anchor.	3 / 3
Trigger Term Quality	The 'Use when the user wants to remix one of their existing ads' trigger surfaces the core keyword ('remix') but offers few natural variations a user might say ('reuse this ad', 'swap the brand', 'make a new version'), so it matches the 'some relevant keywords but missing common variations' anchor rather than broad coverage.	2 / 3
Distinctiveness Conflict Risk	A clear niche — content-goose ad-run folders, source-sample.json, upload-ad-sample/remix-ad consumption — with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (512 lines); consider splitting into references/ and linking	Warning
referenced_paths_exist	Referenced path issues: 6 missing, 2 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: gooseworks-ai/goose-skills
Path: skills/ads/composites/extract-source-sample/SKILL.md
Commit: 8866b2a

Reviewed: 1 day 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.