Agent skill
ideation
Creative ideation for new project concepts. USE WHEN user says "new idea", "what if we built", "I want to create", "brainstorm a project", or discusses ideas without an existing project. Creates IDEA.md and project structure. Not for in-project exploration—use exploration skill for that.
Install this agent skill to your Project
npx add-skill https://github.com/nickpending/momentum/tree/main/skills/ideation
SKILL.md
Ideation
Collaborative creative thinking for new project concepts. Captures insights into IDEA.md and project structure.
Available Environment Variables
These env vars are available in bash commands (use ${VAR} syntax):
${PROJECT_NAME}- Current project name${PROJECT_ROOT}- Current project code directory (e.g.,~/development/projects/argus)${WORKFLOW_PROJECTS}- Obsidian projects root (e.g.,~/obsidian/projects)${OBSIDIAN_DIR}- Obsidian vault root (e.g.,~/obsidian)
Derived paths (placeholders for tool calls):
- Project planning/IDEA.md:
{WORKFLOW_PROJECTS}/{project-name}/ - Explorations:
{OBSIDIAN_DIR}/reference/technical/explorations/ - Later backlog:
{WORKFLOW_PROJECTS}/{project-name}/later.md
Note: The {project-name} references in this skill refer to the project being ideated about (extracted from user conversation), not the current ${PROJECT_NAME} context.
Workflow Decision Tree
User mentions idea/project
↓
Extract project name (ask if ambiguous)
↓
Check for existing {WORKFLOW_PROJECTS}/{project-name}/IDEA.md
↓
├─ Not found → NEW PROJECT flow
├─ Found + major pivot → BIG CHANGES flow
└─ Found + new features → NEW FEATURES flow
↓
Engage in creative discussion
↓
Capture insights mentally during conversation
↓
User triggers save ("save this ideation", "capture this idea", etc.)
↓
Execute save based on detected flow
Examples
Example 1: New project from scratch
User: "I have an idea for a tool that tracks reading habits"
→ Extract project name (ask if unclear)
→ Check for existing IDEA.md → not found → NEW PROJECT flow
→ Engage in creative discussion
→ On "save this ideation" → create IDEA.md
Example 2: Evolving existing vision
User: "I want to pivot argus to focus on home automation"
→ Read existing IDEA.md
→ Discuss major changes → BIG CHANGES flow
→ On save → update IDEA.md, preserve history
Example 3: Adding features to backlog
User: "What if we added dark mode to the dashboard?"
→ Read existing IDEA.md
→ Discuss feature → NEW FEATURES flow
→ On save → append to later.md with ID
Phase 1: Project Detection
BEFORE engaging in discussion, determine the context:
Extract Project Name
Identify the project name from the conversation. If ambiguous or not mentioned, ask directly.
Check for Existing Project
Look for {WORKFLOW_PROJECTS}/{project-name}/IDEA.md using the Read tool.
Classify Scope
Based on findings:
- NEW PROJECT: No IDEA.md exists, completely new concept
- BIG CHANGES: IDEA.md exists, but major pivot/vision shifts being discussed
- NEW FEATURES: IDEA.md exists, discussing additions/improvements
If IDEA.md exists, read it to understand the current vision before proceeding.
Phase 2: Creative Discussion
Approach: Think WITH them, not for them. Build on excitement, challenge assumptions, stay concrete.
Track mentally (don't call out explicitly):
- Core problem, target users, key features
- Success criteria, constraints, unique value
For existing projects: Reference current IDEA.md, build on existing vision For new projects: Focus on problem/solution fit, keep asking "who" and "why"
Phase 3: Saving Ideation
Trigger Phrases
Execute save immediately when user says:
- "save this ideation"
- "let's capture this idea"
- "document this concept"
- "create the project"
Save Execution by Scope
NEW PROJECT
- Create directory:
{WORKFLOW_PROJECTS}/{project-name}/ - Read
references/idea_template.mdfrom this skill - Generate new IDEA.md by filling template with discussion details
- Write to
{WORKFLOW_PROJECTS}/{project-name}/IDEA.md - Confirm: "Created project '{project-name}' at {path}"
- If PROJECT is "workspace": Say "Run
momentum {project-name}to start building."
Template Filling Guidelines:
- Replace
[Project Name]with actual project name - Fill bracketed placeholders with concrete details from discussion
- Use specific examples mentioned during conversation
- Leave sections empty if not discussed (don't invent content)
- Preserve the template structure and section headings
BIG CHANGES
- Read existing
{WORKFLOW_PROJECTS}/{project-name}/IDEA.md - Update with new vision while preserving relevant existing parts
- Move superseded information to "Learning and Evolution" section
- Write updated content back to same location
- Confirm: "Updated vision for '{project-name}'"
Preservation Guidelines:
- Keep any "Learning and Evolution" entries
- Preserve success metrics and constraints that still apply
- Update "Evolution Notes" to document the pivot
- Maintain built features in status section
NEW FEATURES
- Check if
{WORKFLOW_PROJECTS}/{project-name}/later.mdexists - Generate unique ID using
scripts/generate_id.pyfor each feature - Format as:
- idea:: {description} id::{generated-id} captured::{today's date in YYYY-MM-DD format} - Append to
later.md(create file if needed) - Confirm: "Added feature ideas to '{project-name}' backlog"
ID Format: 6-character lowercase alphanumeric (e.g., g7k2m9, x3p5n1)
Feature Description Guidelines:
- Keep descriptions concise but specific
- Focus on user value, not implementation
- Capture enough context to recall the discussion later
- One line per feature idea
Standards
- Use concrete details from conversation, not invented content
- Execute on trigger phrases—don't ask "would you like me to save?"
- Preserve discussion energy in outputs
Resources
references/idea_template.md
Complete template for creating new project IDEA.md files. Read this file when executing NEW PROJECT saves to ensure proper structure and all required sections.
scripts/generate_id.py
Python script that generates 6-character lowercase alphanumeric IDs for feature tracking in later.md files. Execute without loading into context:
python scripts/generate_id.py
Returns format: a1b2c3 (6 random chars from [a-z0-9])
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
exploration
Technical exploration within existing projects. USE WHEN user says "think through", "explore options", "investigate", "how should we approach", or needs to evaluate approaches before implementation. Creates exploration documents in project's explorations/ folder. Not for new project ideas—use ideation skill for that.
obsidian-vault
Search, create, and manage notes in the Obsidian vault with wikilinks and index notes. Use when user wants to find, create, or organize notes in Obsidian.
scaffold-exercises
Create exercise directory structures with sections, problems, solutions, and explainers that pass linting. Use when user wants to scaffold exercises, create exercise stubs, or set up a new course section.
setup-pre-commit
Set up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo. Use when user wants to add pre-commit hooks, set up Husky, configure lint-staged, or add commit-time formatting/typechecking/testing.
git-guardrails-claude-code
Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.
handoff
Compact the current conversation into a handoff document for another agent to pick up.
Didn't find tool you were looking for?