vision
Create, review, or revise a concise project vision document that captures what a project is, who it is for, why it exists, success criteria, constraints, non-goals, and decision principles. Use when starting a new project, clarifying product direction, aligning a codebase for future agent work, defining a north star, or turning a vague idea into docs/vision.md.
What this skill does
<tool_restrictions>
# MANDATORY Tool Restrictions
## REQUIRED TOOLS:
- **`AskUserQuestion`** — Preserve the one-question-at-a-time interaction pattern for every question in this skill, including mode selection, missing context, and draft validation. In Claude Code, use the tool. In Codex, ask one concise plain-text question at a time unless a structured question tool is actually available in the current mode. Keep context before the question to 2-3 sentences max, and do not narrate missing tools or fallbacks to the user.
</tool_restrictions>
<arc_runtime>
This workflow requires the full Arc bundle, not a prompts-only install.
Paths in this skill use these conventions:
- `agents/...`, `references/...`, `disciplines/...`, `templates/...`, `scripts/...`, `rules/...`, `skills/<name>/...` are Arc-owned files at the plugin root. Resolve the plugin root from this skill's filesystem location — it's the directory containing `agents/` and `skills/`.
- `./...` is local to this skill's directory.
- `.ruler/...`, `docs/...`, `src/...`, or any project-relative path refers to the user's project repository.
</arc_runtime>
# Vision Workflow
Create, review, or revise a project vision document. The output should be useful to future humans and agents: specific enough to guide decisions, short enough to be read, and honest about constraints.
## Start
When invoked:
1. State that you are using `/arc:vision`.
2. Determine whether the user wants to create, review, or revise a vision.
3. Inspect existing project context before asking questions.
4. Ask one focused question at a time only when the direction is still unclear.
## Context To Inspect
Read what exists from this list. Do not fail if a file is absent:
- `docs/vision.md`
- `docs/arc/vision.md`
- `README.md`
- `AGENTS.md`
- `CONTEXT.md`
- `docs/brand-system.md`
- `docs/design-context.md`
- `package.json`
- app, package, or domain folder names that reveal the product shape
If `docs/vision.md` exists and the user did not specify a mode, ask:
```
AskUserQuestion:
question: "I found an existing vision document. What would you like to do?"
header: "Existing Vision"
options:
- label: "Review"
description: "Assess the current vision and suggest improvements without overwriting it"
- label: "Revise"
description: "Update the vision based on current context or a new direction"
- label: "Start fresh"
description: "Replace it with a new vision document"
```
If no existing vision exists and the user's intent is clear, draft from available context. Do not force a long interview.
## Useful Questions
Ask only the questions needed to fill real gaps:
- What is the project?
- Who is it for?
- What problem does it solve?
- What should be true if it succeeds?
- What should it deliberately not become?
- What constraints should future work respect?
- What tradeoffs should future decisions optimize for?
## Output
Create or update `docs/vision.md` by default unless the user specifies another path.
Use this structure unless the project already has a better local convention:
```markdown
# Vision
## What This Is
## Who It Serves
## Why It Exists
## Success
## Principles
## Non-Goals
## Open Questions
```
A good Arc vision document normally fits in 500-900 words.
## Writing Rules
- Describe the actual product, audience, and value, not a generic category.
- Prefer concrete nouns and decision language over marketing claims.
- Include non-goals. A vision without boundaries is not operational.
- Include decision principles that can resolve future tradeoffs.
- Separate facts from assumptions when the codebase or user input does not fully support a claim.
- Avoid hype phrases such as "revolutionary", "seamless", "cutting-edge", or "delightful" unless the product context proves they are precise.
- Do not invent business metrics, customer segments, compliance requirements, or committed timelines.
- Do not add broad documentation files beyond the requested vision artifact.
- Keep open questions explicit when direction depends on unresolved assumptions.
## Review Mode
When reviewing an existing vision, lead with issues:
1. Missing or vague audience.
2. Unclear problem statement.
3. No observable success criteria.
4. No non-goals or constraints.
5. Claims contradicted by the codebase.
6. Language too generic to guide implementation.
7. Decision principles that are slogans rather than usable tradeoff rules.
If the user asked only for a review, do not overwrite the file. Provide findings and a proposed revision excerpt instead.
## Revise Mode
When revising:
1. Preserve accurate, specific claims from the existing document.
2. Replace generic or outdated claims with codebase-backed language.
3. Mark assumptions rather than presenting them as facts.
4. Keep the document concise even when the project is complex.
5. Save the final revision to `docs/vision.md` unless the user specified another path.
## Save
When creating or revising:
```bash
mkdir -p docs
# Write to docs/vision.md unless another path was requested.
```
Commit the vision document when the user has asked Arc to own the change or when continuing an Arc workflow where committing is expected:
```bash
git add docs/vision.md
git commit -m "docs: update project vision"
```
## Interop
- `/arc:ideate` reads vision for product and scope context.
- `/arc:launch` checks whether launch posture still matches the project intent.
<completion_check>
Before finishing, verify that the vision:
- Names the product and audience.
- Explains why the project exists.
- Gives future agents enough context to make implementation decisions.
- Includes non-goals or constraints.
- Includes decision principles that can resolve tradeoffs.
- Lists open questions when the vision depends on unresolved assumptions.
</completion_check>
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.