gooseworks-ai/goose-skills / drafts/marketing/learn-my-brand
learn-my-brand
VerifiedLearn my brand once and stop asking me every time. Use when someone says learn my brand, set up my brand, here's our brand book, remember how we talk, what can we claim, can we say this, approve our claims, sign off our voice, our price or product changed, the client approved the claims, pick up where we left off, or check our claims are still current. Reads what GooseWorks already found about the brand, asks only the gaps as drafts to confirm, audits every claim (including ones research marked approved) and gets the ones that matter signed off with proof, scope and review date, turns the voice into rules the owner confirms, checks logo, colours and fonts, and saves it all as it goes so every other job loads it instead of asking. Ends with an unknowns list and a read-back. Not for who our customers are and what they say (use research-audience) or writing positioning from scratch (not covered yet).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 14.5 KB |
| atoms/brain-write-back/ATOM.md | markdown | 6.8 KB |
| atoms/claims-audit/ATOM.md | markdown | 3.9 KB |
| atoms/voice-calibrate/ATOM.md | markdown | 3.1 KB |
| references/brain-map.md | markdown | 10.2 KB |
| references/brand-record-template.md | markdown | 6.2 KB |
| references/category-rules.md | markdown | 9.5 KB |
| references/claims-craft.md | markdown | 13.4 KB |
| references/intake-craft.md | markdown | 7.8 KB |
| references/voice-craft.md | markdown | 6.4 KB |
| skill.meta.json | config | 4.5 KB |
| tests/test-by-hand.md | markdown | 12.0 KB |
atoms/claims-audit/ATOM.md
View on GitHub---
name: claims-audit
description: >-
Internal step of learn-my-brand (hidden). Turns every claim-shaped line the brand brain holds into a
classified candidate with a recommended verdict, and runs the owner's sign-off by risk tier. Used in full
runs, update runs, one-claim checks the owner brings here, client sign-offs and expiry sweeps. The span checks
are the same as the shared ../../../../skills/claims-gate/SKILL.md §2; this is the whole-brand version.
---
# claims-audit
Load `../../references/claims-craft.md` first, and `../../references/category-rules.md` when the category is regulated.
## Input
- The brand's id and category (from learn-my-brand step 3).
- Mode: all claims (full run), the changed claims (update), one named line (one-claim request), or claims past their review date (sweep).
## Procedure
1. **Collect.** From `gooseworks:brand_read` kit (full): `approvedClaims`, `proofPoints`, `valueProps`, `tagline`, `description`, `audience`, `creativeAxes`, the knowledge `positioning`, `existingAds` hooks, offers and angles, site lines quoted inside `voice`; product names, descriptions and offers (`product_ids` in batches); `gooseworks:knowledge_search` `source_types: ["claim"]`; the existing brand record's claims tables; anything the owner's documents state as fact. In one-claim mode, only the line asked about plus any approved claim it resembles.
2. **De-duplicate.** One candidate per distinct statement; keep every place it appears (kit, product, ad).
3. **Classify.** Type and risk from the taxonomy. Mark implied claims (names, visuals the owner describes, testimonials implying typical results).
4. **Find proof.** `knowledge_search` `source_types: ["document", "evidence"]` with the claim as the query (snippets only point to a document; read the document itself before calling it proof). Record proof strength. A press quote needs its article URL and date; a rating needs platform, count and capture date; an award needs body, year and category.
5. **Recommend a verdict:** keep / keep with qualifier (write it) / reword (write the safer wording) / needs proof (say exactly what would unlock it) / cut. Fill every field of the claim record except approver and date.
6. **Draft the never-say list** from the rule pack, the platforms the brand uses, and anything the owner has said; each with a safe alternative and a source tag.
7. **Triage, then run the sign-off by risk** (claims-craft.md §4): low-risk batch; medium in visible groups of 3–5; high one at a time for the five or so with most reach; every other high-risk line recorded as not approved yet without asking. Switch to the product-facts-only default the moment the owner signals time pressure. Say once that this is not legal advice. Ask the not-public question once.
8. **Record each answer** with the approver's words verbatim and the date. If the person answering is not the approver, the new lines go to the record's "Decisions needed" sheet to forward (the claims status becomes `pending-client` only if nothing is approved yet; an approved list stays `signed-off`); still record facts they state and pull claims they say are false. In client sign-off mode, map the pasted reply to the sheet's rows and record it with provenance.
## Output (to `atoms/brain-write-back`, which records it in the exact claims-gate §5 format)
- Claims the approver approved, with complete records.
- Candidates not approved: wording, why, what would unlock it, safe alternative.
- Never-say entries with source and alternative; not-public items with dates.
- Value props, tagline, description and audience rewritten so every factual line is an approved claim (for the owner's yes).
- The approver's verbatim words per decision.
## Rules
- A brain label of `approved` is not a sign-off. Only the approver's answer in this step is.
- Never strengthen a claim when rewording. Never invent proof, a study, a number or a customer.
- A claim without proof stays out, however good it sounds.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |