Skip to content
Agentmart
Submit a tool

product-discovery

Run Product Manager discovery as either a coordinator routine that creates self-contained project subtasks or a project-specific subtask that performs deep…

Skill by alvarosanchez

IndexedMIT★ 1
Write a review

Chalk Labs

Need help deploying product-discovery?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/alvarosanchez-product-discovery

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Run Product Manager discovery as either a coordinator routine that creates self-contained project subtasks or a project-specific subtask that performs deep review and may create one QA-assigned backlog product issue.

Use cases: testing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/product-discovery

What's inside

Skill name
product-discovery
Description
Run Product Manager discovery as either a coordinator routine that creates self-contained project subtasks or a project-specific subtask that performs deep review and may create one QA-assigned backlog product issue.
License
MIT
Instructions
94 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Product Discovery

Use this skill whenever Product Manager is handling monthly-product-discovery or a project-specific product-discovery child issue.

## Modes

First decide which mode you are in:

- Coordinator mode: the parent monthly-product-discovery routine issue asks you to create or reuse project-specific product-discovery child issues or subtasks.
- Project subtask mode: a project-owned product-discovery child issue or subtask asks you to do the deep review for one repository.

The parent routine issue is coordination only. Do not perform deep repository review, market research, competitor research, candidate selection, or product development issue creation from the routine issue itself.

## Coordinator Mode

In the parent routine issue:

- Inspect active Paperclip projects and include only active Micronaut-related projects that map to managed GitHub repositories.
- Exclude internal company-operating projects such as `company-operations`.
- Exclude `micronaut-projects/micronaut-project-template`; it is a repository template and file sync source, not an actual Micronaut project.
- Before creating a child issue or subtask, search for an existing open or already-created product-discovery child issue or subtask for the same routine issue and project. Also search for orphan or top-level product-discovery issues for the same project from recent routine attempts. Reuse, update, or reparent the existing issue when possible; if it cannot be safely reparented, record a blocker instead of creating another duplicate.
- Create at most one product-discovery child issue or subtask per eligible project for a routine run.
- Put each child issue or subtask in the actual corresponding Paperclip project, set `parentId` to the routine issue when supported, and set assignee to Product Manager (`product-manager`).
- Give each child issue a stable title such as `Product discovery: <project name> (<YYYY-MM-DD>)` unless an existing project-specific product-discovery subtask already exists.
- The child issue description must be self-contained and complete. It must say to use the product-discovery skill and include the project repository, the discovery checklist, duplicate-check expectations, output sections, and the allowed outcomes. Do not rely on the child agent reading the parent routine description.
- Do not create top-level product development issues from the parent routine issue.

The routine report should list every eligible project, every skipped project and skip reason, every reused or created child issue URL, and blockers that prevented child issue creation.

## Project Subtask Mode

In a project-specific product-discovery child issue or subtask:

- Read the child issue body and use this `product-discovery` skill as the source of truth.
- Inspect the repository README, docs, examples, recent releases, current open issues, recently closed feature requests, and synced Paperclip project notes before researching outside the project.
- Inspect previous product-discovery reports, prior discovery run notes, and project subtask reports before proposing a feature, including created product issues, no-create decisions, rejected candidates, and duplicate decisions from earlier runs.
- Research market, competitor frameworks, adjacent technologies, and developer workflows with live web search when useful.
- Summarize the repository's current capabilities before naming a gap.
- Identify candidate feature gaps that are concrete enough for engineering and valuable enough for maintainers to consider.
- Deduplicate candidates against open and closed GitHub issues in the same repository and against existing Paperclip product development issues in the project.
- Do not propose or create the same previously proposed feature candidate from a prior run unless new evidence materially changes the decision. When the same gap appears again, cite the earlier created issue, duplicate finding, rejected candidate, or no-create decision instead of opening another request.
- Pick at most one high-value non-duplicative feature request for the project in a routine run.
- When justified, create one top-level Paperclip product development issue in the same Paperclip project with `status: backlog`, `workMode: standard`, no `parentId`, and assignee QA (`qa-engineer`). Pass the backlog status explicitly so assigned issue creation does not auto-dispatch the proposal as `todo`.
- Include `Intended GitHub label: type: enhancement` in the product development issue body unless an equivalent Paperclip label can be applied safely.
- Do not publish issues to GitHub from Product Manager discovery.

The project subtask report must record the capability summary, research sources, previous run findings, candidate gaps, duplicate checks, selected feature or no-create decision, repeated candidate decisions, created product development issue URL when one exists, and blockers.

## Product Development Issue Body

When a project subtask creates a product development issue, include:

```md
## Problem

Describe the user-visible problem and the affected user persona.

## Market and competitor evidence

- Name the competitor framework, tool, platform, or ecosystem signal.
- Link or cite the source you used.
- Explain the specific capability or workflow users can already get elsewhere.

## Current Micronaut capability

Summarize what this repository already supports, including docs, APIs, configuration, or known issues inspected during discovery.

## Gap

Explain the missing capability and why it matters now.

## Proposed feature

Describe the expected behavior, configuration shape, API shape, documentation shape, or user workflow in enough detail for architecture and engineering to refine it.

## Implementation considerations

Name likely affected modules, docs, tests, compatibility boundaries, release-targeting implications, and security considerations.

## Acceptance criteria

- A user can complete the target workflow without custom glue code.
- The feature has tests or documented verification steps that cover the main success path and at least one failure or compatibility case.
- The user-facing docs explain how to enable, configure, and verify the feature.
- Existing behavior remains backward-compatible unless the issue explicitly asks for a breaking change.

## Prior art and duplicate check

List the GitHub searches, issues, pull requests, docs, releases, existing Paperclip issues, previous product-discovery reports, prior project subtask reports, and prior run created issue or no-create decisions checked before opening this request.

Intended GitHub label: type: enhancement
```

From alvarosanchez/micronaut-agent-company, MIT.

More from alvarosanchez/micronaut-agent-company

Signals

Stars
1
Forks
0
Open issues
1
Last commit (this folder)
Jul 3, 2026 (3 months ago)
Last release
Sep 22, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from alvarosanchez

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get product-discovery running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying product-discovery?

CallGet help
Cite this page

product-discovery skill. Agent Mart. https://agentmart.dev/skills/alvarosanchez-product-discovery (last updated 2026-10-24)