brand-research

Kickoff research for a brand you haven't worked on before — web research, existing-ad analysis from the Meta Ad Library, editorial-grammar profiling, sourced + AI-generated brand assets, hook/CTA libraries, and an ad concept brief. Produces one reusable brand-context pack (brand-summary, visual-identity, competitors, audience, existing-ads, brand-grammar, an asset manifest, and a concept brief) in a single pass. Use when starting on a brand the workspace hasn't touched.

Quality

90%

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

80%

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

The body is dense, concrete, and well-sequenced with strong fallback/decision logic, but it under-uses its own bundle: the output contract is duplicated inline and the deterministic verifier is never invoked in the workflow, costing it both workflow-clarity and progressive-disclosure points.

Suggestions

Add an explicit final workflow step that runs `python scripts/verify_pack.py --brand-dir <output_dir>` as the ship-gate, with a 'fix and re-run on failure' feedback loop — this closes the validation gap and lifts workflow_clarity.

Replace the inline header/manifest duplication (steps 6–7 and the Output section) with a one-line pointer to `references/output-contract.md` for the normative schema, keeping only the overview in SKILL.md.

Reference each composed-atom script where relevant (e.g. link `scripts/render_product_shot.py` from the image-generation step) so the bundle structure is actually signaled from the body.

Dimension	Reasoning	Score
Conciseness	Assumes Claude's competence with no concept-explanation filler — every section (inputs, composed atoms, workflow, decision rules, quality checks, failure modes) is task-specific; a few lines in step 4 restate the grammar purpose, but the body is lean overall and matches the 'every token earns its place' anchor.	3 / 3
Actionability	Gives concrete, copy-paste-ready specifics — exact section headers, exact folder/file paths, the manifest `kind` enum, the `video-research.json` shape, atom names with parameters (`max_ads=max_existing_ads`), and model names (`gpt_image_2`, `nano_banana_2`); per the instruction-only note, the absence of in-body shell commands is not penalized since the guidance is actionable.	3 / 3
Workflow Clarity	A clear 9-step sequence with disambiguation gates and fallbacks, but the obvious validation checkpoint — the bundled `scripts/verify_pack.py` ship-gate — is never wired into the step sequence, so checkpoints read as prose checks rather than an explicit 'run the verifier' feedback loop; not 3 because that checkpoint is missing from the flow.	2 / 3
Progressive Disclosure	Well-sectioned, but the normative output contract is duplicated inline (steps 6–7 + Output restate what `references/output-contract.md` defines) and none of the bundle files (`references/output-contract.md`, `scripts/verify_pack.py`, etc.) are referenced from the body; matches the 'references present but not clearly signaled; content that should be separate is inline' anchor.	2 / 3
	Total	10 / 12 Passed

Description

100%

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 strong: third-person, concrete, and answers both what it does and when to use it with a clear, distinctive trigger. It is somewhat dense and jargon-heavy ('editorial-grammar profiling', 'hook/CTA libraries'), but all terms are task-relevant and grounded.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('web research, existing-ad analysis from the Meta Ad Library, editorial-grammar profiling, sourced + AI-generated brand assets, hook/CTA libraries, and an ad concept brief') and enumerates every pack component, matching the 'lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers both what (the research actions + pack outputs) and when ('Use when starting on a brand the workspace hasn't touched'), satisfying the explicit-trigger anchor; not 2 because the 'Use when' clause is present.	3 / 3
Trigger Term Quality	Includes natural terms a user starting brand work would say ('brand', 'research', 'ads', 'Meta Ad Library', 'assets', 'concept brief', 'starting on a brand the workspace hasn't touched'), giving good coverage of natural trigger phrasing.	3 / 3
Distinctiveness Conflict Risk	Clear niche — kickoff research producing a brand-context pack, triggered by 'a brand the workspace hasn't touched' — with distinct triggers unlikely to fire for generic ad/creative skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: gooseworks-ai/goose-skills
Path: skills/ads/composites/brand-research/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.