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 skills add OutlineDriven/outline-driven-development --skill artifact-arenagit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/outlinedriven/outline-driven-development/artifact-arena)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/artifact-arena"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/artifact-arena.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.01164 |
| Opus 5 | $0.00017 | $0.00582 |
| Sonnet 5 | $0.00007 | $0.00233 |
| Haiku 4.5 | $0.00003 | $0.00116 |
Grade A, and why
artifact-arena 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 today.
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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Artifact arena
Contract
| Field | Bound contract |
|---|---|
| Trigger | User runs /artifact-arena to generate and judge competing artifact implementations. |
| Authority | Reversible local: writes only local candidate worktrees and the synthesis artifact; rollback is discarding candidate worktrees and deleting the synthesis note. No remote mutation. |
| Side effect | Spawns isolated candidate workers, each writing to its own local path. No remote, credential, VCS, or published mutation. |
| Done | One verified synthesis with provenance and rejection reasons. |
Inputs
- Task prompt (required): the artifact each candidate must produce, stated as a single contract.
- Candidate count N (required): how many parallel candidates to spawn. Minimum 2.
- Model pool (optional): models for candidates and the cross-judge. When absent, use available models, preferring a different family for the cross-judge than the parent.
- Shared grounding (optional): a path to context every candidate reads before producing.
Procedure
-
Frame. State the artifact each candidate produces. Derive a rubric of 3-6 concrete gradeable criteria; each criterion names an observable behavior, not a vague quality. The rubric is the picker's tool; candidates see only the task prompt, not the rubric. Assign each candidate its own output path (a git worktree where possible, otherwise a unique local directory). N candidates sharing one path is shared mutable state and corrupts results. Done when: rubric of 3-6 criteria is derived and each candidate has its own output path.
-
Fan out. Spawn all N candidate workers in one batch, each receiving the task prompt, the shared grounding path if supplied, its own output path, and instructions to produce both the artifact and a short rationale naming the alternatives it considered and what it rejected. The rationale is mandatory; without it, grafting in step 5 is unreliable. If a candidate produces no output, proceed with N-1 and record the dropout. Done when: all N candidates are spawned and each produces an artifact with rationale or is recorded as a dropout.
What ships with it
1 file 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.
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.
- today Changed · -11 tokens per session 78661da8273d
- 2d ago First seen · 47 lines · 46 tokens per session scan A de9cb0f18cf0
artifact-arena is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed yesterday), licensed Apache-2.0. It adds 35 tokens to every session and 1,164 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 2026-09-03.
Other skills, from other repositories
audit-project
Run an iterative multi-agent code audit until critical and high findings are resolved. Use when the user says "audit my code", "find all the bugs", "deep code audit", "iterative review", or "review until clean".
duet
Use when the user invokes /duet, says "pair with me", or faces aesthetic, architectural, or irreversible decisions.
goal-prompt-drafting
Use when asked to draft copy-ready /goal objectives for long-running agents; returns one normalized one-line objective with measurable end state, grounded proof, easy-out invariants, a stop clause, and a Missing list. Not for source or remote-system changes.
handoff-prompt
Use when the user asks for a handoff, delegation, or clipboard-ready prompt for another agent: a standalone path-free prompt copied to the clipboard, confirmed by title. Not for session-snapshot briefs — use handoff; never remote, credential, publish, deploy, or irreversible.
drill
Use when a concept needs practising rather than explaining: run a scaffolded exercise from worked example to independent problem, quiz the learner, run spaced recall over what they cleared, or probe for the gaps blocking what they want next. For explanation, use explain-concept; for an end-to-end build, use capstone.
snippet-image-rendering
Use when the user explicitly names snipgrapher and wants code rendered to a polished PNG, SVG, or WebP at an explicit local path. Probes the installed CLI for supported flags and writes only the named image file. Not for other renderers, publishing, or remote actions.