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/defaultperson/agent-setup/ultrathinknpx skills add DefaultPerson/agent-setup --skill ultrathinkgit clone --depth 1 https://github.com/DefaultPerson/agent-setupWrote 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/defaultperson/agent-setup/ultrathink)<a href="https://agentmods.dev/skills/defaultperson/agent-setup/ultrathink"><img src="https://agentmods.dev/badge/skills/defaultperson/agent-setup/ultrathink.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 | $0.00043 | $0.00219 |
| Opus 5 | $0.00022 | $0.00110 |
| Sonnet 5 | $0.00009 | $0.00044 |
| Haiku 4.5 | $0.00004 | $0.00022 |
Grade A, and why
ultrathink 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 4d 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.
What it actually says
Use the maximum amount of ultrathink. Take all the time you need.
Reasoning
- Exhaustive step-by-step reasoning. Explicitly check every assumption — do not guess.
- Before concluding, list what could be wrong with your reasoning. Stress-test edge cases.
- If uncertain — state assumptions, propose how to verify quickly.
Execution
- Parallelize independent work: launch multiple agents for research, exploration, or verification where tasks don't depend on each other.
- Prefer depth over breadth: fully solve one subproblem before moving to the next.
- After completing work, re-read modified files and verify correctness.
Output
- Short TL;DR first, then a concrete numbered step-by-step solution.
$ARGUMENTS
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.
- 4d ago First seen · 25 lines · 43 tokens per session scan A 09068e1ce3e5
ultrathink is a skill published in the GitHub repository DefaultPerson/agent-setup (11 stars, last pushed 2mo ago), licensed MIT. It adds 43 tokens to every session and 219 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 2026-08-30.
Other skills, from other repositories
bark-notify-test
Send a Bark verification notification for the current Claude Code Bark setup on macOS or Windows. Use when the user wants to test delivery, confirm encryption works, or check whether the configured completion notification is healthy.
bark-notify
Use this skill whenever the user wants Claude Code to notify an iPhone through Bark when a task, turn, or coding session finishes. Works on macOS, Linux, and Windows. Trigger on requests about Bark notifications, Claude completion alerts, iPhone push reminders, completion messages, Bark hook setup, or Bark…
bark-notify-setup
Set up Bark completion notifications for Claude Code on macOS or Windows by safely writing the local Bark sender script and merging hooks into /.claude/settings.json. Use when the user explicitly asks to configure Bark for them or to directly install the completion notification flow.
bark-notify-uninstall
Remove Bark notification configuration from Claude Code on macOS, Linux, or Windows. Use when the user wants to uninstall, remove, or clean up Bark notifications, delete the Bark hook, or reset their Bark setup.
address-pr-review
Critically triage pull request review comments against PR requirements and codebase reality, implement only valid fixes, and reply only when declining a suggestion. Use when the user asks to address PR comments, review feedback, CodeRabbit/Bugbot threads, or /address-pr-review.
triage-agent-eval-failures
Triage failing @novu/agent-evals scenarios to decide whether a failure is real or flaky, and whether to fix the playbook/prompt or the test (grader, tape, scenario, or judge). Use when an agent-evals scenario fails, when the user asks why an eval is red, or when deciding whether to fix the test or the prompt.