Skip to content
Agentmart
Submit a tool

facilitating-product-vision

Facilitates an owner-led product vision session and produces two artifacts: a strategy-only vision document plus roadmap items (outcome themes under…

Skill by hierovision

IndexedMIT★ 1
Write a review

Chalk Labs

Need help deploying facilitating-product-vision?

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/hierovision-facilitating-product-vision

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

Facilitates an owner-led product vision session and produces two artifacts: a strategy-only vision document plus roadmap items (outcome themes under Now/Next/Later, emitted as ROADMAP.md rows per the sibling item-row schema). Use when the user says 'define/refresh the product vision', 'our roadmap is too lean', 'turn the vision into slices', 'run a vision/strategy session', or 'help me shape the product direction'. Not for: ranking an existing backlog (triaging-requirements), planning one named item (designing-architecture), persisting a roadmap as issues (managing-github-issues), refining a single issue's ACs (refining-issue-acceptance), or read-only product analysis of a draft (council-product). The phrase 'product vision' is the trigger — this skill is about product/strategy, not visual UI critique; UI/screenshot 'vision' reviews belong to council-ux/validating-ui.

Use cases: documents, design

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/facilitating-product-vision

What's inside

Skill name
facilitating-product-vision
Description
Facilitates an owner-led product vision session and produces two artifacts: a strategy-only vision document plus roadmap items (outcome themes under Now/Next/Later, emitted as ROADMAP.md rows per the sibling item-row schema). Use when the user says 'define/refresh the product vision', 'our roadmap is too lean', 'turn the vision into slices', 'run a vision/strategy session', or 'help me shape the product direction'. Not for: ranking an existing backlog (triaging-requirements), planning one named item (designing-architecture), persisting a roadmap as issues (managing-github-issues), refining a single issue's ACs (refining-issue-acceptance), or read-only product analysis of a draft (council-product). The phrase 'product vision' is the trigger — this skill is about product/strategy, not visual UI critique; UI/screenshot 'vision' reviews belong to council-ux/validating-ui.
License
MIT
Instructions
373 lines in SKILL.md · view on GitHub

Files (8)

  • SKILL.md
  • evals.json · evals/
  • items.md · evals/fixtures/backlog/
  • ROADMAP.md · evals/fixtures/lean-roadmap/
  • owner-notes.md · evals/fixtures/lean-roadmap/
  • README.md · evals/fixtures/single-item/
  • artifact-format.md · references/
  • consumer-session.md · references/

SKILL.md (first lines)

# Facilitating Product Vision

Run a structured, owner-led session that turns product ambition into a
strategy-only vision artifact plus roadmap items. The output is two
artifacts (vision document + provisional `ROADMAP.md` rows) plus a
working session record — never a rebuild of an existing roadmap and
never a strategy invented under the owner's name.

The session is owner-driven. Every judgment call is owner-owned and
recorded with a one-line rationale; the facilitator does not invent
strategy. The artifact is the contract: a separate reviewer reads it
cold and must be able to defend the strategy and roadmap-item set.

## Contents

- The session (process checklist)
- Step 1 — Inputs & constraints (incl. Step 1a — decide the session mode)
- Step 2 — Strategy batch + Checkpoint 1
- Step 3 — Roadmap batch + Checkpoint 2
- Step 4 — Compliance & phase gates + Checkpoint 3
- Step 5 — Write the artifacts
- Step 6 — Council consults and handoffs
- Boundary routing
- References

## The session (process checklist)

Copy this checklist and check off items as you complete them. Owner
checkpoints are batched into three decision points — not one pause per
piece — to keep the session conversational and avoid late-session
rubber-stamping.

```
Vision Session Progress:
- [ ] 1. Inputs & constraints (ambition, current assets, envelope, non-negotiables, vision-artifact home)
- [ ] 1a. Decide mode (interactive multi-turn | declared one-shot | decisions pre-supplied inline) — the mode determines what fills Open Questions and what the final response must say
- [ ] 2. Strategy batch draft (vision, segments+JTBD, alternatives+non-goals+anti-vision, value model, north star + guardrails)
- [ ] 2a. CHECKPOINT 1 — strategy batch: present draft as questions-with-proposals; owner decides, every call recorded (skipped only when decisions are pre-supplied — recorded instead)
- [ ] 3. Roadmap batch draft (3–6 outcome themes banded Now/Next/Later under the capacity envelope; each theme's work emitted as roadmap items in the sibling item-row schema — ACs NOT emitted, they are plan-time artifacts; risk-first ordering)
- [ ] 3a. CHECKPOINT 2 — roadmap batch: same Q-with-proposal cadence (skipped only when decisions are pre-supplied — recorded instead)
- [ ] 4. Compliance & phase gates draft (consent/privacy/regulatory milestones, phase gates that unlock later segments)
- [ ] 4a. CHECKPOINT 3 — compliance + gates: confirm gates and owners (skipped only when decisions are pre-supplied — recorded instead)
- [ ] 5. Write artifacts (vision doc at the owner-confirmed home using EXACT Title-Case headings; append outcome-theme + roadmap-item rows to ROADMAP.md per the sibling item-row schema; record the session as a working artifact)
- [ ] 6. Optional council consults + handoffs (triaging-requirements ranks, designing-architecture plans the top roadmap item)
```

## Step 1 — Inputs & constraints

Capture what the strategy is constrained by. Read the consumer's
existing decision-doc convention (an `AGENTS.md`, a `docs/`, an ADR
index) for the project name, the team's pre-existing context, and the
canonical home for a long-lived document.

Ask the owner, and record in a one-line recap:

- **Ambition** — what the product is for, in the owner's own words.
- **Current assets** — what already exists (live code, design system,
  installed skills, an established `ROADMAP.md`, a prior vision doc).
- **Budget / time envelope** — team capacity, regulatory deadlines,
  quarters the session is planning into.
- **Non-negotiables** — pre-decided constraints the strategy must
  respect (legal posture, a platform commitment, a partner contract).

Then ask the **canonical home for the vision artifact** and record the
owner's answer. The vision artifact's path is a consumer decision, not
a default. The skill cites the artifact's content contract
(`references/artifact-format.md`, resolved against this skill's own
directory — not the project's); the owner decides where it lives.

If the owner names an established `ROADMAP.md`, read it. The session
appends to it; it never rebuilds it from scratch (the
established-once rule below).

### Step 1a — Decide the session mode

Three modes, picked from the prompt before drafting starts. The
mode determines what fills the artifact and what the final response
is required to say. Pick one and stay in it for the whole session:

- **Interactive multi-turn (default).** The session pauses at each
  of the three checkpoints and asks the owner. Owner decisions are
  recorded in-session as they are given.
- **Declared one-shot (no back-and-forth).** The owner has said
  they cannot answer back (e.g., "this is a one-shot, I'll read the
  doc later"). The session still writes the artifact (it does not
  end with an intention to write it later). Sections that require
  owner input — Segments & Jobs-to-be-Done, Alternatives &
  Non-goals, Value & Business Model, North Star + Guardrails,
  Compliance & Phase Gates, the capacity envelope — get a
  placeholder of the form
  `Open question: <what is needed, which section it will land in>`
  rather than invented content. The final response addresses the
  owner and lists every open question in plain text — using the
  literal words "owner" and "question" — so the consumer can
  recover the artifact on a later pass. The session does not
  present an invented strategy as decided.
- **Decisions pre-supplied inline.** The owner has already supplied
  every (or most) owner decisions for THIS run in this turn's
  prompt, or in the owner's own notes file the session has read. Each
  pre-supplied decision is recorded as a decision with its source
  cited. Anything the owner has NOT pre-supplied is surfaced as an
  Open Question. The session does not invent the missing inputs.

The boundary between *interactive* and *pre-supplied* is whether
the owner can answer back; the boundary between *one-shot* and
*pre-supplied* is whether the owner has already given the
decisions. A prompt that says "complete in one pass, here are the
answers" is *pre-supplied*, not one-shot: the decisions exist;
they are recorded. A prompt that says "I can't answer, write what
you can" is one-shot: the decisions do not exist; they are
surfaced.

## Step 2 — Strategy batch + Checkpoint 1

Draft the strategy batch. Read
[references/artifact-format.md](references/artifact-format.md) (resolved
against this skill's own directory) for the fixed section order and
section schemas — the body cites it, never duplicates it. Cover:

From hierovision/ai-framework, MIT.

More from hierovision/ai-framework

Signals

Stars
1
Forks
0
Open issues
7
Last commit (this folder)
Oct 12, 2026 (19 days ago)

From GitHub (fetched Oct 24, 2026).

Details

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

More from hierovision

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 facilitating-product-vision 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 facilitating-product-vision?

CallGet help
Cite this page

facilitating-product-vision skill. Agent Mart. https://agentmart.dev/skills/hierovision-facilitating-product-vision (last updated 2026-10-24)