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 sixb-ai/sixb --skill sixb-artifactsgit clone --depth 1 https://github.com/sixb-ai/sixbWrote 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/sixb-ai/sixb/sixb-artifacts)<a href="https://agentmods.dev/skills/sixb-ai/sixb/sixb-artifacts"><img src="https://agentmods.dev/badge/skills/sixb-ai/sixb/sixb-artifacts/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sixb-ai/sixb/sixb-artifacts"><img src="https://agentmods.dev/badge/skills/sixb-ai/sixb/sixb-artifacts.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00047 | $0.01359 |
| Opus 5.5 | $0.00019 | $0.00544 |
| Sonnet 5.5 | $0.00009 | $0.00272 |
| Haiku 4.5 | $0.00005 | $0.00136 |
Grade C, and why
sixb-artifacts scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf work/q3-review && cp -r "$SIXB_ARTIFACTS_DIR/art_k3m9q2x7tp" work/q3-review How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Artifacts
An artifact is a deliverable the user opens beside the conversation. It has a kind, a title, and versions that stay readable. Publish one when the user asks for a report, a document, a table, or a file, or when the answer is long-form work or data they will read, reuse, or share. When a few sentences answer the question, answer in the chat.
Kinds
| Kind | Publish it for | Read first |
|---|---|---|
document |
A report, brief, memo, plan, or write-up: anything read top to bottom | document.md |
sheet |
Tabular data the reader sorts, sums, charts, or downloads: metrics, lists, trackers, exports | sheet.md |
slides |
A deck to present: a review, a pitch, a plan for a meeting | slides.md |
page |
A small web app the reader works with: a calculator, a planner, a dashboard with filters | page.md |
canvas |
Screens side by side with arrows, a brainstorm or retro of notes, a diagram | canvas.md |
file |
One file in a format the user asked for (PDF, DOCX, XLSX, PPTX, CSV), or a binary | file.md |
To show or link to another artifact, such as a sheet's chart in a report, read references.md.
Data from the project's objects goes in a query, so the artifact stays live: read queries.md.
Prefer a document to a DOCX or PDF, a sheet to an XLSX or CSV file, and slides to a PPTX,
unless the user asked for that format: they render in the app, and their versions stay readable side by side. A report
with a table of a dozen rows is a document; data the reader works with is a sheet.
Publish
-
Write the artifact's files in a working directory, such as
work/q3-review/. -
Call
publish_artifactwith that directory (or one file), the kind, and a title:publish_artifact({ path: "work/q3-review", kind: "document", title: "Q3 operations review" }) → Published "Q3 operations review" as artifact art_k3m9q2x7tp, version 1. -
If it refuses, nothing was published. Fix what the message names and publish again. If you cannot, tell the user the deliverable could not be published; never describe it as ready.
-
Fix what each warning names and publish its next version when it matters to the reader. A deck also comes back with a screenshot of each slide, a page with one at a laptop's width and one at a phone's, and a canvas with one of each board: look at them (
slides.md,page.md,canvas.md, Check). -
In the reply, say in a sentence or two what you published and what stands out. Do not paste its content, a path, or its id as text: the user sees the artifact. To show a chart or a range of it in place, use its address (
references.md). A figure you cite from live data is the one you read, and what the artifact shows moves on as the objects change, even at a pinned version: date it with when you read it, in the user's time zone ("As of Oct 23, 10:42, three invoices are overdue.").
What ships with it
35 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.
- canvas.md 5.5 KB
- document.md 2.8 KB
- examples/canvas/onboarding-flow/boards/first-job.html 1.6 KB
- examples/canvas/onboarding-flow/boards/region.html 1.4 KB
- examples/canvas/onboarding-flow/boards/shared.css 2.6 KB
- examples/canvas/onboarding-flow/boards/welcome.html 1.5 KB
- examples/canvas/onboarding-flow/canvas.json 2.1 KB
- examples/canvas/retro/canvas.json 3.1 KB
- examples/document/incident-report/document.md 1.4 KB
- examples/document/operations-review/document.md 1.5 KB
- examples/document/operations-review/figures/restorations.svg 2.3 KB
- examples/page/collections-desk/index.html 7.7 KB
- examples/page/collections-desk/queries/overdue.json 401 B
- examples/page/dispatch-board/data/dispatch.json 5.4 KB
- examples/page/dispatch-board/index.html 14 KB
- examples/page/fleet-planner/data/plan.json 1.1 KB
- examples/page/fleet-planner/index.html 22 KB
- examples/page/service-explorer/data/service.json 21 KB
- examples/page/service-explorer/index.html 25 KB
- examples/sheet/overdue-invoices/queries/overdue.json 401 B
- examples/sheet/overdue-invoices/workbook.json 865 B
- examples/sheet/parts-inventory/sheets/Inventory.csv 1.6 KB
- examples/sheet/parts-inventory/sheets/Spend.csv 120 B
- examples/sheet/parts-inventory/workbook.json 1.2 KB
- examples/sheet/service-metrics/sheets/Quarters.csv 253 B
- examples/sheet/service-metrics/sheets/Regions.csv 349 B
- examples/sheet/service-metrics/workbook.json 2.2 KB
- examples/slides/launch-plan/index.html 2.5 KB
- examples/slides/q3-review/index.html 6.0 KB
- file.md 920 B
- page.md 11 KB
- queries.md 3.3 KB
- references.md 3.6 KB
- sheet.md 11 KB
- slides.md 6.6 KB
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 · +6 lines f2552cec07a9
- 5d ago Changed · +14 lines c0e8da7997a0
- 9d ago First seen · 76 lines · 47 tokens per session scan C 398d2df2a8a0
sixb-artifacts is a skill published in the GitHub repository sixb-ai/sixb (126 stars, last pushed today), licensed MIT. It adds 47 tokens to every session and 1,359 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-19.
Other skills, from other repositories
google-workspace
Gmail, Calendar, Drive, Docs, Sheets via gws CLI or Python.
box
Box manages cloud files, sharing, search, and metadata.
csv-workbench
Analyze CSV files in /mnt/data and return concise numeric summaries.
review-artifact-comments
Only use when the user explicitly invokes this skill by name.
potpie-source-ingestion
Use when the user explicitly asks to ingest, refresh, or deeply understand a repository, PR, issue, ticket, runbook, incident report, document, or web link into Potpie. The harness performs todo-driven discovery, uses local/GitHub/integration tools and read-only subagents when available, builds evidence-backed…
figures
Makes publication-quality plots from data with matplotlib, learning curves, scaling laws, benchmark and ablation comparisons, Pareto trade-offs, heatmaps and confusion matrices, sized for the page, vector, with uncertainty shown. Use whenever results are plotted, charted or visualized for a paper, report or answer, or…