Bouncer extract-requirements-from-mock
Extract requirements document from HTML mock iteration
install
source · Clone the upstream repo
git clone https://github.com/imbue-ai/bouncer
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/imbue-ai/bouncer "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.claude/skills/extract-requirements-from-mock" ~/.claude/skills/imbue-ai-bouncer-extract-requirements-from-mock && rm -rf "$T"
manifest:
.claude/skills/extract-requirements-from-mock/SKILL.mdsource content
Extract Requirements from Mock
You are extracting a clean requirements document from an HTML mock iteration session.
Input
Feature name: $ARGUMENTS
Steps
-
Locate the feature folder: Look for
docs/$ARGUMENTS/- If it doesn't exist, ask the user to provide the correct feature name
-
Gather all context:
- Read
for the original description, Q&A, and all logged UI tweaksmocks.context.md - Read
to understand the final UI statemocks.html - Review the conversation history for any additional context
- Read
-
Extract requirements: Create
with:docs/$ARGUMENTS/requirements.md- Focus only on user-facing behavior - what the user sees and can do
- No implementation details - don't specify how things should be built
- No acceptance criteria - just describe the behavior
-
Structure the requirements doc:
# <Feature Name> - Requirements This document specifies the user-facing requirements for <feature>. It describes *what* the system must do from the user's perspective. --- ## 1. <First Major Area> - **REQ-XXX-1:** <requirement> - **REQ-XXX-2:** <requirement> ## 2. <Second Major Area> ... -
Present the requirements to the user for review.
-
Optionally regenerate mocks: Ask the user if they would like to:
- Rename the old mocks and regenerate fresh ones based on the new requirements
- Keep the existing mocks as-is