artifact

A skill for publishing documents, images, diagrams, and PDFs as persistent links through an Artifacts server.

In plain words
What is it for?
Use it to publish or update plans, design documents, status reports, diagrams, generated images, and PDFs.
Why use it?
It gives people a stable page to review shared files and lets living documents be updated without changing their link.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/dbtlr/artifacts/artifact
Any agent
npx skills add dbtlr/artifacts --skill artifact
Clone the repo
git clone --depth 1 https://github.com/dbtlr/artifacts

Made for: Claude Code, Codex.

Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,588 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00047 $0.01588
Opus 5 $0.00023 $0.00794
Sonnet 5 $0.00009 $0.00318
Haiku 4.5 $0.00005 $0.00159

Measured yesterday against content hash c28471c59e91, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

artifact 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 yesterday.

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.

skills/artifact/SKILL.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Publish a document or file to the Artifacts server and hand back the returned URL. The same URL keeps working across edits, so living artifacts should be updated in place rather than re-created.

Why This Skill Exists

Documents produced mid-task — a plan, a design doc, a status update, a diagram — are easiest to review as a rendered page with a link, especially when working with the user remotely. Artifacts gives every such document a stable URL; this skill teaches the decision an agent has to make before writing one: update the existing artifact, or create a new one, and with which template and metadata.

Core Workflow

Step 1: Look for an existing artifact or collection first

Before creating anything, call list_artifacts (pass project if you know it) to check whether it already exists. For a multi-artifact result such as a document with generated images, call list_collections and filter list_artifacts by the chosen collection.

  • Update beats duplicate for living documents. A plan, status report, or design doc that's still evolving should be updated via update_artifact, not recreated — the URL stays the same across updates, so anyone holding the old link keeps seeing current content.
  • Create only when genuinely new — a document that hasn't existed before, or a distinct artifact from ones already listed (a new phase's plan is not the same artifact as the previous phase's).
  • When updating, content is a full replacement, not a diff. If you're editing rather than regenerating, call get_artifact with includeContent: true, edit the result, then send the full replacement to update_artifact.

Step 2: Choose metadata

  • title — short and human-scannable; it's what shows up in list views. ("Q3 Migration Plan", not "Plan for migrating the thing we discussed.")
  • project — a stable string per repo or initiative, reused across every artifact in that body of work (matches what list_artifacts filters on in Step 1). Pick it once and keep it consistent so Step 1 keeps finding the right artifacts.
  • description — one line, written for the list view, not the document body.
  • collection — optional shared metadata for related independent artifacts. Reuse one value for a document and its images so another agent can rediscover the set. A collection does not own its members.

Read the full file on GitHub · 94 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. yesterday First seen · 94 lines · 47 tokens per session scan A c28471c59e91

Subscribe to this mod's changes

artifact is a skill published in the GitHub repository dbtlr/artifacts (0 stars, last pushed 11d ago), licensed MIT. It adds 47 tokens to every session and 1,588 once invoked, about $0.0002 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 2026-08-31.