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 tatwo214/TATWO-ULTRAWORK-beta --skill tatwo-ultraworkgit clone --depth 1 https://github.com/tatwo214/TATWO-ULTRAWORK-betaWrote 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/tatwo214/tatwo-ultrawork-beta/tatwo-ultrawork)<a href="https://agentmods.dev/skills/tatwo214/tatwo-ultrawork-beta/tatwo-ultrawork"><img src="https://agentmods.dev/badge/skills/tatwo214/tatwo-ultrawork-beta/tatwo-ultrawork/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/tatwo214/tatwo-ultrawork-beta/tatwo-ultrawork"><img src="https://agentmods.dev/badge/skills/tatwo214/tatwo-ultrawork-beta/tatwo-ultrawork.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 | $0.00028 | $0.00704 |
| Opus 5 | $0.00014 | $0.00352 |
| Sonnet 5 | $0.00006 | $0.00141 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
tatwo-ultrawork 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 8d 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TATWO Ultrawork — Public Portable Edition
Activation
Use when the user explicitly asks for TATWO Ultrawork, multi-engine rooms, lead/worker/reviewer coordination, or a goal-first collaboration loop.
State the actual surface:
Public workflow activewhen only this skill is available.Runtime connectedonly when an independently configured host adapter is verifiably reachable.- Never imply that this skill grants process, filesystem, network, credential, deployment, or device authority.
Workflow
- Goal — restate the desired outcome, success evidence, non-goals, constraints, and protected surfaces.
- Context — inspect the smallest relevant evidence. Separate known facts, assumptions, contradictions, and inaccessible facts.
- Plan — name the files or surfaces to touch, verification method, rollback, and human gates. Do not operationalize before context is sufficient.
- Loops — use the fewest bounded workers needed. Every room receives a title, role, working root, brief, acceptance evidence, and stop condition.
- Review — a different identity critiques the result. Review is advisory and cannot self-approve.
- Verify — the lead reruns applicable checks, inspects primary artifacts, compares each success criterion, and reports missing evidence.
- Close — distinguish completed, blocked, failed, skipped, and unverified work.
Safety defaults
- Preview and read-only discovery come first.
- Provider adapters, credentials, sessions, host mutation, network access, installations, background services, deployment, device control, secrets, destructive actions, and live financial actions are outside this public skill.
- Never discover or inherit another provider's configuration automatically.
- Reject paths outside the explicitly allowed working root; treat symlinks as a separate trust decision.
- Do not place secrets, personal paths, account identifiers, browser state, private prompts, real receipts, or real project fixtures into shareable artifacts.
- A process being alive, a worker saying “done,” a generated report, or merged report text is not acceptance evidence.
- If the runtime or authority boundary is unclear, stop at a plan or preview and label the missing proof.
What ships with it
3 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.
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.
- 8d ago First seen · 66 lines · 28 tokens per session scan A 3b40857a368c
tatwo-ultrawork is a skill published in the GitHub repository tatwo214/TATWO-ULTRAWORK-beta (0 stars, last pushed 8d ago), licensed Apache-2.0. It adds 28 tokens to every session and 704 once invoked, about $0.0001 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 2026-09-04.
Other skills, from other repositories
orchestrate
This is the extended reference for the headless T0 orchestrator. The condensed identity is in agents/orchestrator/CLAUDE.md.
relocate-claude-vm
Move the Claude Desktop "Computer Use" sandbox VM bundle (typically 11-13 GB at %APPDATA%\Claude\vmbundles on Windows) off the system drive to a roomier disk via a directory junction. Transparent to Claude Desktop, reversible. Use when the system drive is low on space and a disk scan shows vmbundles as a top consumer.…
collaborating-with-gemini
Delegates coding tasks to Gemini CLI for prototyping, debugging, and code review. Use when needing algorithm implementation, bug analysis, or code quality feedback. Supports multi-turn sessions via SESSIONID.
report-generator
Multi-AI collaborative report generator. Uses Claude for planning/writing and Gemini for diagram generation. Creates professional reports with auto-generated figures from your research content.
sparc-methodology
SPARC (Specification, Pseudocode, Architecture, Refinement, Completion) comprehensive development methodology with multi-agent orchestration.
stream-chain
Stream-JSON chaining for multi-agent pipelines, data transformation, and sequential workflows.