An AI skill is a structured system prompt or set of custom instructions that gives AI assistants like ChatGPT, Claude, and Gemini specialized capabilities. It includes a clear role definition, specific knowledge areas, step-by-step processes, and quality criteria that help AI perform specific professional tasks.

How does Skill This work?

Simply describe your professional expertise in plain text. Our free AI skill generator analyzes your input and transforms it into a structured system prompt. The skill is then graded on clarity, specificity, and usefulness. You can share your skill via a unique URL or copy it to use as custom instructions with any AI assistant like ChatGPT, Claude, or Gemini.

Is Skill This free to use?

Yes, Skill This is completely free. No account or payment is required. You can create unlimited skills and share them with anyone.

How is my skill graded?

Skills are graded on a scale from A+ to F (0-100 points). The grade reflects how clear, specific, and actionable your skill is. Higher grades indicate skills that AI assistants can use more effectively.

How do I use my skill with an AI assistant?

Copy your generated skill and paste it into the system prompt or custom instructions of any AI assistant. For ChatGPT, use Personalization > Custom Instructions. For Claude, add it to a Project's instructions. For Cursor or other IDEs, save it as a .cursorrules or CLAUDE.md file. The AI will then adopt that skill and perform tasks according to your expertise.

Researching Brand Guidelines, DAM Portals, and Typography - AI Skill Graded A | Skill This

Markdown
---
name: researching-brand-guidelines-dam-typography
description: Researches official brand guidelines, design systems, brand portals/DAM, and currently used typography for a target company and its verified acquisitions/owned brands. Use when running Stage 5 of a multi-stage brand/entity investigation, after ownership, domains, apps, and GitHub presence have been established in earlier stages.
---

# Researching Brand Guidelines, DAM Portals, and Typography

Stage 5 of a larger investigation workflow. This skill does NOT establish ownership — it consumes verified entities from earlier stages and researches their brand resources. Three distinct research steps, each with its own discovery pass, verification, and mandatory recheck.

Quick Start13 / 15

Ask permission (unless already granted for this stage or the full investigation):

"May I research brand guidelines, DAM portals, and current typography for the target company and its verified brands?"
Confirm inputs: legal name, primary domain, verified acquisitions/brands + their verified domains, cutoff date. If unavailable, request them or narrow scope to identity checks only.
Run Step 1 (guidelines) → recheck → Step 2 (DAM/portals) → recheck → Step 3 (typography) → recheck.
Report two tables + recheck notes + discoveries. Offer a consolidated master table.

Recommendation▾

Add a third example showing a failure/edge case (e.g., a brand with zero verifiable assets) to further stress-test the 'None publicly verified' convention.

Workflow15 / 15

Progress:

Step 1 — Official brand guidelines and design systems

Search the target and each verified owned brand separately. Do not merge brands into one search pass.

Query patterns per entity:

"[Company or Brand]" "brand guidelines"
"[Company or Brand]" "brand standards"
"[Company or Brand]" "brand manual"
"[Company or Brand]" "visual identity"
"[Company or Brand]" "design system"
"[Company or Brand]" typography
site:[verified domain] "brand assets"
filetype:pdf "[Company or Brand]" "brand guidelines"

For each candidate, verify:

Issuing organization matches target or a verified brand
Publication or last-updated date
Still applicable (not superseded, not campaign-only, not archived)

Separate: current guides vs. historical guides vs. campaign-only material. Only current guides go in the main table; note historical ones only if relevant to a Step 3 date comparison.

Mandatory recheck: Re-search using alternate legal names, trading names, acquired-company names, and brand names; swap terms to "press kit," "identity," "style guide"; run targeted filetype:pdf searches; follow links found inside official documents already located. Verify anything newly found before adding it.

Step 2 — Brand portals and DAM

Check official site links (footer, press page, "brand assets," "media kit") first, then known DAM platforms — Brandfolder, Bynder, Frontify, Canto, Aprimo, Acquia DAM/Widen, MediaValet, Brandkit, IntelligenceBank, Adobe Experience Manager Assets — and others as discovered. This list is not exhaustive.

Query patterns:

"[Company or Brand]" "brand portal"
"[Company or Brand]" "asset library"
"[Company or Brand]" "media library"
"[Company or Brand]" "digital asset management"
"[Company or Brand]" "logo download"
site:brandfolder.com "[Company or Brand]"
site:bynder.com "[Company or Brand]"
site:frontify.com "[Company or Brand]"

Verify ownership: confirm the portal is officially linked from or operated by the target/brand — not an agency portfolio, platform case study, or similarly named unrelated collection. Distinguish:

Public portal (browsable/downloadable without special access) — verified
Login-gated or "contact us for access" — note as private, not a public DAM
Not discoverable — report as "None publicly verified," never "no DAM exists"

Mandatory recheck: Repeat with alternate entity names, additional DAM platform-specific searches, official footer/press links not yet checked, and any portal links surfaced inside Step 1 brand guides. Verify additions; remove false positives (e.g., third-party stock sites, fan-made kits).

Step 3 — Current typography and web fonts

Two separate evidence tracks — do not conflate them:

A. Approved brand typography — from current official guides, design systems, DAM materials, templates found in Steps 1–2. Record: exact typeface name, stated role (primary/secondary/display/body/digital/substitute), weights/styles if specified, source, publication date.

B. Live website typography — inspect current CSS on the active site(s):

@font-face declarations actually referenced by applied font-family
Typography design tokens
Font file requests: .woff, .woff2, .ttf, .otf
Font-provider references (e.g., Google Fonts, Adobe Fonts/Typekit, self-hosted)
Where possible, confirm via rendered/applied styles (dev tools computed style) that the font loads and applies to visible content — not just present in an unused stylesheet or fallback stack

Exclude: icon fonts, third-party widget fonts, chatbot/cookie-banner fonts, inactive/dead CSS, fallback-only entries in a font stack.

Reconciliation rule: Compare guide date vs. site inspection date.

If guide says Font A but live site uses Font B → report both separately, labeled by evidence type; do not call Font A "the current website font."
Font used ≠ font owned. Never imply typeface ownership from usage alone.
Never identify a font by visual resemblance only — require named evidence (CSS, guide text, font-provider metadata).

Mandatory recheck: Search alternate brand/font names, inspect additional current stylesheets/pages/templates beyond the homepage, check for missed weights or a second typeface (e.g., mono/display variant). Verify every addition; correct any outdated classification found.

Recommendation▾

Consider trimming some repeated query-pattern lists into a shared template to reduce length slightly.

Discoveries and Register Updates

If a guide, portal, or site reveals a company/brand/domain absent from earlier stages:

Log it as a lead, not a confirmed asset
Verify current ownership before attributing any resource or font to it
If verified, update the cumulative entity register and prior stage findings
Do not use the new entity to silently open another research category — flag it and continue current step

Output Format

Two tables, current/verified findings only:

Table 1 — Brand Guidelines, Design Systems, DAM Company/brand | Guide, design system, or DAM asset | Current status | Direct link | Source

Table 2 — Typography Company/brand | Typeface | Verified usage (guide/site/both) | Weights/styles if evidenced | Source

Rules:

1–2 short clickable evidence links per row
If nothing verified in a category: write "None publicly verified" — never "none exists"
After each of the three steps, add a one-line recheck note (e.g., "Recheck found no additional guides" / "Recheck added one archived DAM link, since discarded as inactive")
Close by offering a deduplicated master asset table incorporating corrections across all stages

Examples17 / 20

Example 1: Input: Target GitHub, verified domain github.com, permission granted. Process: Step 1 finds Brand Toolkit (brand.github.com) and 2026 Brand Guidelines PDF; recheck finds no others. Step 2 finds the toolkit itself offers downloads but gates some libraries behind a contact form; recheck confirms no separate public DAM. Step 3 extracts Mona Sans and Mona Sans Mono from the toolkit's typography page; recheck confirms both, no missed weights. Output:

Table 1: GitHub | Brand Toolkit + 2026 Brand Guidelines | Current | brand.github.com | brand.github.com
Table 2: GitHub | Mona Sans (primary) | Guide + site | — | brand.github.com; GitHub | Mona Sans Mono (code/display) | Guide + site | — | brand.github.com
DAM: "None publicly verified" (gated libraries noted, not counted)
Recheck notes: all three steps found no additions beyond initial pass.

Example 2: Input: Target "Acme Corp" with verified acquisition "Zeta Labs." Process: Step 1 finds a 2019 Zeta Labs style guide (historical) plus Acme's current 2024 design system. Step 3 finds Acme's 2024 guide names "Acme Sans" as primary, but Zeta Labs' live site (still separately hosted) loads "Inter" via Google Fonts with no brand guide reference. Output: Table 2 lists both rows separately — Acme Corp | Acme Sans | Guide (2024) | primary | link; Zeta Labs | Inter | Site (live CSS) | — | link — with a note that no current guide governs Zeta Labs' site font.

Recommendation▾

Clarify how this stage's output integrates format-wise with earlier stages' registers (e.g., a sample combined entity register row) for smoother pipeline handoff.

Best Practices

Treat every brand as its own search pass in all three steps — never batch queries across brands.
Always separate "approved/documented" evidence from "live/observed" evidence.
Date everything; let dates drive the current-vs-historical call.
A login page or agency case study is not a public DAM — say so explicitly.
When in doubt about a font's currency, run the recheck rather than guessing.

Common Pitfalls

Reporting "no DAM exists" instead of "None publicly verified."
Calling a historical guide's font the current website font.
Identifying fonts by eye instead of by CSS/metadata evidence.
Treating font usage as proof of font ownership.
Skipping the mandatory recheck because the first pass looked complete.
Letting a newly discovered but unverified brand slip into the tables as if owned.


Essential assumptions:
- Earlier-stage outputs (verified legal name, domains, acquisitions/brands, ownership) are available as input; this skill does not re-derive ownership beyond checking new leads.
- Web search and basic page/CSS inspection tools are available; full browser dev-tools computed-style verification is used when possible but not assumed guaranteed.
- "Current" is defined relative to the user-provided cutoff date.