discovery-research-wasp-drone — Claude Code subagent: what it does
discovery-research-wasp-drone runs the continuous product discovery cycle: outcome scoping, an Opportunity Solution Tree, JTBD interview scripts and prototype experiments.
AGPL-3.0-or-later ★ 96 updated 2026-10-18
/plugin marketplace add legioncodeinc/vibe-coding-tools
/plugin install business-and-product-tools@wasp-nestWhat discovery-research-wasp-drone does
The subagent builds and maintains an Opportunity Solution Tree mapping a desired outcome to opportunity clusters, solutions and experiments. For a chosen opportunity it drafts a five-act JTBD interview script, scores a solution's assumptions on importance versus uncertainty, and designs the smallest test for the riskiest one, whether that's a paper mock, Wizard of Oz, concierge, fake door or landing page.
Before any of that it must resolve the paired skill discovery-research-stinger through hooks/resolve-component.py and read its SKILL.md first; the procedure does not start otherwise. Output lands under library/discovery/: the stated outcome, the tree, dated interview scripts, assumption maps and experiment plans.
Usability testing of shipped features, UI decisions and PRD writing stay with other agents (quality-wasp-drone, ux-ui-wasp-drone, library-wasp-drone). It can weigh pre-build evidence against stated assumptions but won't certify market demand, and it won't run contacts, spend or publication without separate authorization.
What it helps with
- pin down a single measurable discovery outcome
- build an Opportunity Solution Tree from interview data
- draft a JTBD interview script for a target opportunity
- score a solution's assumptions by importance and uncertainty
- design a minimal prototype experiment for the riskiest assumption
Example prompts for the agent
- Run a discovery session and help define the outcome
- Build an Opportunity Solution Tree for this opportunity
- Write a JTBD interview script for this opportunity
How to install discovery-research-wasp-drone
1Plugin marketplace
/plugin marketplace add legioncodeinc/vibe-coding-tools/plugin install business-and-product-tools@wasp-nest2Manually: put the files in place
.claude/agents/discovery-research-wasp-drone.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/discovery-research-wasp-drone.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/discovery-research-wasp-drone.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about discovery-research-wasp-drone
How do I install discovery-research-wasp-drone in Claude Code?
Add the marketplace with /plugin marketplace add legioncodeinc/vibe-coding-tools, then install the plugin with /plugin install business-and-product-tools@wasp-nest.
What does discovery-research-wasp-drone do?
It runs the full discovery loop: defining an outcome, building an Opportunity Solution Tree, scripting JTBD interviews, mapping assumptions and designing prototype experiments.
How does it differ from other agents in this suite?
It operates before implementation and hands off to agents like library-wasp-drone once an outcome is agreed and an opportunity is chosen; it leaves shipped-feature usability testing, UI decisions and PRD writing to separate agents.