Forge-core DebrandPrompt

Remove hardcoded vendor names, brand references, and external-service dependencies from a prompt-shaped document. USE WHEN an adopted skill hardcodes specific tool names, company brands, or paid service dependencies that would bias the skill toward one ecosystem.

install
source · Clone the upstream repo
git clone https://github.com/N4M3Z/forge-core
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/N4M3Z/forge-core "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/DebrandPrompt" ~/.claude/skills/n4m3z-forge-core-debrandprompt && rm -rf "$T"
manifest: skills/DebrandPrompt/SKILL.md
source content

DebrandPrompt

Strip vendor and brand references from a prompt-shaped document so the skill generalizes across ecosystems. Referenced by ForgeAdopt as the

debrand
transform.

What to remove

PatternExampleReplace with
Hardcoded vendor tool name in core guidance"Use Docker Scout for vulnerability scanning""Use a vulnerability scanner"
"Powered by X" or partner-credit language"Built on top of Supabase"Remove
External-service dependency as mandatory"Configure Kong as the API gateway""Configure an API gateway"
Specific author attribution in body"The codex team recommends..."Remove attribution, keep the recommendation
Marketing affiliate language"With built-in Auth0 integration""With authentication integration"
First-person voice tied to a specific org"Our team uses X"Describe the pattern, not the org

What to preserve

  • Vendor names that appear in anti-pattern citations: "avoid Inter, Roboto, Arial" stays — the names are the signal
  • Vendor names that identify a standard (HTTP, OAuth, SAML, TLS) — these are protocol names, not brands
  • Commands and package names in code blocks when they are the actual invocation the user needs
  • License attributions in frontmatter or companion files

The anti-pattern test

Ask: if I replaced this vendor name with

<category>
, does the sentence still convey the instruction?

  • "Use Docker Scout" → "Use a vulnerability scanner" — passes, debrand it
  • "Avoid Inter and Roboto" → "Avoid <font>s and <font>s" — fails, keep the names because they ARE the content

Procedure

  1. Grep for common vendor markers: capitalized product names,
    @
    handles, URLs pointing to non-standard services.
  2. For each match, apply the anti-pattern test.
  3. When removing, replace with the category (scanner, gateway, font, service) — do not delete in place.
  4. Update
    allowed-tools
    if a removed vendor tool implied a tool grant.
  5. Re-read; the skill should read as ecosystem-neutral where appropriate.

Constraints

  • Never remove vendor names from anti-pattern lists — they are the directive
  • Do not neutralize protocol names (HTTP, TLS, OAuth) — these are universal
  • If the skill is explicitly about a specific vendor (for example, a Fakturoid skill), do not debrand — the vendor is the scope
  • When in doubt between debranding and preserving, preserve and flag for review