Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/howells/arc/visionnpx skills add howells/arc --skill visiongit clone --depth 1 https://github.com/howells/arcWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00076 | $0.02140 |
| Opus 5 | $0.00038 | $0.01070 |
| Sonnet 5 | $0.00015 | $0.00428 |
| Haiku 4.5 | $0.00008 | $0.00214 |
Grade A, and why
vision scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<tool_restrictions>
Ask one question at a time. In Claude Code use AskUserQuestion; elsewhere ask a single concise plain-text question. Keep any lead-in to 2-3 sentences. Don't narrate missing tools or fallbacks.
</tool_restrictions>
<arc_runtime>
Requires the full Arc bundle. Arc-owned paths (agents/, references/, disciplines/, templates/, scripts/, rules/, skills/) resolve from the plugin root — the directory containing agents/ and skills/. Everything else is the user's repository.
</arc_runtime>
Vision Workflow
Create or maintain a project's root CONTEXT.md: its goals, product boundary, non-goals, decision principles, and domain language. 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:
- State that you are using
/arc:vision. - Determine which mode applies to the project's CONTEXT.md: review, extend, or start fresh.
- Inspect existing project context before asking questions.
- 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:
CONTEXT.mdCONTEXT-MAP.mddocs/vision.md(legacy — fold its content into CONTEXT.md if found)README.mdAGENTS.mddocs/brand-system.mddocs/design-context.mdpackage.json- app, package, or domain folder names that reveal the product shape
If CONTEXT.md already exists, read it first and preserve the user's content. Update and extend it — sharpen goals, boundary, and glossary — rather than overwriting. If the user did not specify a mode, ask:
AskUserQuestion:
question: "I found an existing CONTEXT.md. What would you like to do?"
header: "Existing Context"
options:
- label: "Review"
description: "Assess the current CONTEXT.md and suggest improvements without overwriting it"
- label: "Extend"
description: "Update and sharpen goals, boundary, and glossary, preserving existing content"
- label: "Start fresh"
description: "Replace it with a new CONTEXT.md"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 229 lines · 76 tokens per session scan A 46d0720039bf
vision is a skill published in the GitHub repository howells/arc (25 stars, last pushed 18d ago), licensed MIT. It adds 76 tokens to every session and 2,140 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
n8n-agents
Design n8n AI agents the right way. Use when building or editing any @n8n/n8n-nodes-langchain. AI node — an AI Agent, LLM chain, Text Classifier, or Information Extractor — and whenever the user mentions AI agents, LLM with tools, tool calling, $fromAI, system prompts, agent memory, sessionId, structured/JSON output…
n8n-error-handling
Wire n8n error handling so failures are loud, structured, and recoverable. Use when building any webhook/API workflow, a scheduled or unattended workflow, or any path where a silent failure would drop user-visible work — and whenever the user mentions error handling, onError, continueErrorOutput, error…
n8n-subworkflows
Build reusable, composable n8n sub-workflows. Use when extracting shared logic, building anything multi-step or reused across workflows, or any workflow over 10 nodes — and whenever the user mentions sub-workflows, Execute Workflow, reuse, shared/common logic, modular workflows, "Define Below" inputs…
n8n-validation-expert
Interpret validation errors and guide fixing them. Use when encountering validation errors, validation warnings, false positives, operator structure issues, or need help understanding validation results. Also use when asking about validation profiles, error types, the validation loop process, or auto-fix capabilities.…
n8n-binary-and-data
Handle files and binary data in n8n correctly. Use when working with files, images, PDFs, attachments, uploads or downloads, base64, vision/multimodal input, or when an AI agent needs a file as tool input or output — and whenever the user mentions $binary, binaryPropertyName, "read the PDF", "attach the file", "send…
n8n-code-tool
Write JavaScript or Python for the n8n Custom Code Tool (@n8n/n8n-nodes-langchain.toolCode) — the AI-agent-callable tool, NOT the workflow Code node. Use when building a Code Tool attached to an AI Agent, writing code that an LLM will invoke, parsing the query input, returning a string result, defining an input schema…