agentleFS
itsaspacestationitsaspacestation/natural-intelligence/skills/discovery/SKILL.md

discovery

Skill2 starsChanged 3 days agoApache-2.0View on GitHub

Use when the user proposes a feature or product idea with no plan yet, asks whether something is worth building, to challenge a feature, compare with existing products or competitors, run a market scan, or write an opportunity brief. Also use when the user mentions product discovery, build versus buy, PR-FAQ, or jobs to be done.

28 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use specify, brainstorm and 4 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ArvorCoSkill

    specify

    Create feature specification from natural language description. Use when starting a new feature or creating a spec. Triggers on: create spec, specify feature, new feature spec.

    Also in 3 projects that use this

  • nicolasestremSkill

    brainstorm

    Structured idea generation and solution exploration with systematic analysis.

    Also in 4 projects that use this

  • Hugging FaceSkill

    implement

    Make tests pass. Invoke after /write-tests produces failing tests.

    Also in 6 projects that use this

  • MicrosoftSkill

    init

    Also in 3 projects that use this

  • AnthropicSkill

    architecture

    Create or evaluate an architecture decision record (ADR). Use when choosing between technologies (e.g., Kafka vs SQS), documenting a design decision with trade-offs and consequences, reviewing a system design proposal, or designing a new component from requirements and constraints.

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 6 projects that use this

The file

Files
SKILL.md

Discovery

When to use

  • User proposes a feature or product idea and no workspace or plan exists
  • User asks whether something is worth building, or to challenge a feature
  • User asks what already exists: competitors, alternatives, build versus buy
  • User asks for an opportunity brief, a PR-FAQ, or a market scan

Overview

Discovery decides whether and what to build. Planning decides how. The job is killing risk before engineering commits, so a day spent deciding not to build is a win, never a failure. The output is a one-page opportunity brief ending in a verdict: build, buy, test first, or drop. On build, the plan skill takes over.

Reference: market-scan.md. Read it at workflow step 3.

Workflow

  1. Reframe as an outcome. Rewrite the request as the user or business outcome it serves, measurable in one sentence. A feature with no stated outcome is the first red flag: ask one question, or in a delegated context derive the outcome from the brief and record it as an assumption.
  2. Challenge the four big risks (Cagan, Inspired): value (will anyone use it), usability (can they figure it out), feasibility (can we build it in the time we have), viability (does it work for the business: cost, legal, support). One sentence per risk, each marked low, open, or blocking, with its evidence.
  3. Scan the market. Browse what already solves the job: competitors, adjacent products, open source, internal tools, and doing nothing. Search the job in the end user's words before any solution-shaped query, or adjacent players stay invisible. Compare by how well each completes the user's job, never by feature checklist. Query shapes, table format, and sources live in market-scan.md.
  4. Map opportunities before solutions (Torres, opportunity solution tree). One outcome per tree: outcome, then unmet needs, then candidate solutions, then assumption tests. Hold at least three candidates before choosing, and "integrate an existing product" from step 3 is always one of them.
  5. Test the riskiest assumption cheaply. Name the single assumption that kills the idea fastest. Name the cheapest test: a query over existing data, a prototype, five user conversations. Never a build.
  6. Write the brief to docs/discovery/<name>.md and stop for the verdict. In a delegated context, decide, record the reason, and continue.

The opportunity brief

One page. Every claim cites a source: a user quote, a metric, a ticket, or a URL.

SectionHolds
OutcomeThe measurable user or business outcome, one sentence
EvidenceDemand signals, each with its source
MarketFull player list (rejected ones included), scan verdict, and the comparison table from step 3
RisksThe four risks, each low, open, or blocking, with why
VerdictBuild, buy, test first, or drop, with the reason
Kill criteriaThe evidence that would reverse the verdict

Not: "Competitor X has dashboards, we need dashboards." Yes: "Users export to Excel weekly to answer one question (tickets #312, #340); X answers it in-product."

Boundaries

This skill decides whether and what, never how. On a build verdict it hands the brief to the plan triage; tasks, ADRs, and code start there. Weighing conflicting reviews or field reports belongs to bias-analysis. Claims about the existing codebase follow evidence-based-analysis.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.