Forge — Brand Styleguide

Prompts as source code: versioned, testable, owned by the team that wrote them.

Color

Dark workshop canvas; mint pass-signal; per-provider hues
Canvas
#0E1014
Canvas 2
#161A21
Canvas 3
#1E232C
Rule
#262C36
Brand / Pass
#7CF0BD
Brand Deep
#2BBF85
Accent (model B)
#6EA8FE
Accent (model C)
#C792EA
Amber / Regression
#F0B86E
Danger
#FF6B6B
Ink 100
#E2E6EC
Ink 300
#9BA3B0

Type

System humanist sans for chrome; mono where data lives
Run it against three models. Pick the one that passes.

v14 regressed on Claude — reverted to v12

Side-by-side evaluation

Pack: Onboarding email rewrites

Variables in this prompt

A prompt is a unit of work. It has a body, typed variables, parameters, a history of versions, and a small set of tests that travel with it. Run it against any provider you've connected. The output is captured so you can compare what was good last week against what's good today.

Last edited 7m ago by Devon T. — version 14 created on save.

model = "claude-3.5-sonnet"   temperature = 0.2   {{tone}}

Buttons

Mint primary = "Run" — the most-pressed control
Default Hover No key set

Forms

Focus, error, mono textarea for prompt bodies
Appears in the library and in pack listings.
Must be a positive integer.
Run on save Notify pack subscribers on publish Stream output as it generates Wait for full response
Public listing enabled Allow forks

Cards

Prompt card with body preview and variable chips

Cold outreach — short

v143 casesEdited 7m ago
You are writing to {{persona}}.
Keep it under 90 words.
Open with an observation about {{company}}.
Pass 3/3 claude gpt-4.1

Summarize support ticket

v68 casesEdited 2d ago
Summarize the conversation below in 3 bullets.
Last bullet must name the customer's blocker.

Conversation:
{{transcript}}
Regressed on gpt-4.1 Internal

Eval table

Cases × provider/model grid
Case claude-3.5-sonnet gpt-4.1 llama-3.1-70b Avg cost
Greets by first name PassPassPass $0.0021
Mentions the company's recent product PassFailPass $0.0024
Stays under 90 words PassPassFail $0.0019
Does not use the word "synergy" PassPassPass $0.0018

Badges & pills

Pass Regression Fail claude-3.5-sonnet gpt-4.1 Internal v14 +3 cases

Alert & toast

This run will cost about $1.84.
You're at 72% of the daily budget. The owner set $50/day for this workspace.
Saved as version 15 Pack published — 3 buyers notified

Empty state

No fishing for clicks — say what to do next

No prompts here yet

Save your first prompt to start the version history. Or import a pack from the marketplace.

Voice

A prompt is a unit of work. Save it. Diff it. Run it against the models you actually use. The library remembers what worked last week so you can argue with what's working today. Forge — voice sample