.agents/skills/oat-idea-ideate/SKILL.md
Use when continuing brainstorming for an existing idea or starting from a scratchpad entry. Guides conversational ideation and refinement.
npx skillsauth add tkstang/open-agent-toolkit oat-idea-ideateInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Resume brainstorming on an existing idea, or pick one from the scratchpad to start exploring.
OAT MODE: Ideation
Purpose: Lightweight brainstorming through natural, exploratory conversation. Capture the essence of an idea without formal structure or implementation detail.
BLOCKED Activities:
ALLOWED Activities:
Self-Correction Protocol: If you catch yourself:
Recovery:
Print a phase banner once at start using horizontal separators, e.g.:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ OAT ▸ IDEATE [project] ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Replace [project] with [global] when operating at user level.
Before multi-step work, print step indicators, e.g.:
[1/3] Resolving active idea…[2/3] Loading discovery document…[3/3] Starting brainstorming session…Determine whether to operate at project level or user (global) level.
Resolution order:
$ARGUMENTS contains --global → use user level.oat/config.local.json has activeIdea pointing to a valid directory → use project level~/.oat/config.json has activeIdea pointing to a valid directory → use user level.oat/ideas/ AND ~/.oat/ideas/ exist →
ask: "Ideas exist at both project and user level. Where should this idea go?"
options: "Project (.oat/ideas/)" / "Global (~/.oat/ideas/)".oat/ideas/ exists → use project level~/.oat/ideas/ exists → use user levelSet variables:
| Variable | Project Level | User Level |
| ---------------- | ---------------------- | ------------------------ |
| IDEAS_ROOT | .oat/ideas | ~/.oat/ideas |
| TEMPLATES_ROOT | .oat/templates/ideas | ~/.oat/templates/ideas |
Read activeIdea from config:
IDEA_PATH=$(oat config get activeIdea 2>/dev/null || true)
If valid (directory exists with discovery.md):
If missing or invalid:
ls -d {IDEAS_ROOT}/*/ 2>/dev/null
{IDEAS_ROOT}/scratchpad.md and extract unchecked entries (- [ ]), including any nested notes for contextoat config set activeIdea "{idea-path}" and proceed to Step 2oat-idea-new skill (.agents/skills/oat-idea-new/SKILL.md) and executing its Steps 3-7 (Initialize Ideas Directory, Scaffold Discovery Document, Update Backlog, Check Scratchpad, Set Active Idea Pointer). Then proceed to Step 2 with the new idea.oat-idea-new skill to create one, or run the oat-idea-scratchpad skill to capture a quick idea seed." Then stop.Read {IDEAS_ROOT}/{idea-name}/discovery.md.
Show the user a brief summary of the current state:
If oat_idea_state: summarized:
brainstorming, continue to Step 4summary.md, then stopAdd a new session header to the Notes & Discussion section:
### Session: YYYY-MM-DD
Update frontmatter: oat_idea_last_updated: YYYY-MM-DD
Enter a free-form conversational flow:
When the user signals they're done (or the conversation reaches a natural pause):
oat_idea_last_updated in frontmatterdiscovery.mdoat-idea-ideate skill to pick up where you left off"oat-idea-summarize skill to finalize this idea"oat-idea-ideate skill and pick a different idea"oat-idea-scratchpad skill to jot down a new idea seed"oat_idea_last_updated reflects today's datedocumentation
Use when OAT implementation changes and repository reference docs must be synchronized. Updates .oat/repo/reference to match current behavior.
business
Merge multiple analysis artifacts into a single coherent report with provenance tracking. Reads existing artifacts from /deep-research, /analyze, and /compare.
testing
Use when the user questions or suspects an agent claim is wrong. Adversarially gathers evidence to verify or refute the claim using the best sources available in the current environment.
tools
Use when prioritizing backlog work or evaluating a roadmap. Produces value-effort ratings, dependency mapping, and execution recommendations.