Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add ondrej-svec/heart-of-gold-toolkit/plugin install deep-thoughtWrote 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/ondrej-svec/heart-of-gold-toolkit/think)<a href="https://agentmods.dev/skills/ondrej-svec/heart-of-gold-toolkit/think"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/think/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/ondrej-svec/heart-of-gold-toolkit/think"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/think.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 | $0.00081 | $0.02104 |
| Opus 5.5 | $0.00032 | $0.00842 |
| Sonnet 5.5 | $0.00016 | $0.00421 |
| Haiku 4.5 | $0.00008 | $0.00210 |
Grade A, and why
think 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 2mo 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Think
Deep reasoning for decisions that matter. Expert perspectives, devil's advocate, what-if scenarios, and structured tradeoff analysis.
Boundaries
This skill MAY: read code/docs for context, reason, analyze, present conclusions. This skill MAY NOT: edit code, create files, run commands, implement anything, deploy.
This is thinking, not doing. Present the analysis — the user decides what to act on.
Common Rationalizations
| Shortcut | Why It Fails | The Cost |
|---|---|---|
| "Default to Expert Panel for everything" | Not always the right lens. A security decision needs Devil's Advocate. | Generic advice that misses the specific risk |
| "High confidence — the analysis is thorough" | Most decisions are genuinely medium confidence. | Over-commitment to a choice that should have been hedged |
| "Skip the counter-argument" | If you can't articulate why the alternative might be better, you don't understand the tradeoff. | Blind spot becomes the failure mode |
| "The first framing is fine" | How you frame the question determines what answers you see. | Solving the wrong problem with rigor |
Phase 0: Frame the Question
Entry: User invoked /think with a question, topic, or nothing.
If invoked with a question (e.g., /think should we use WebSockets or SSE?):
- Use the question as starting point
- Gather relevant context: what does the codebase currently do? What are the constraints?
If invoked without a question: Use AskUserQuestion (header: "Topic", question: "What decision or problem do you want to think through?") with contextual options if possible, otherwise let the user type via the automatic "Other" option.
If invoked with ultrathink:
- Enable extended thinking — deeper analysis, more perspectives, longer reasoning chains
Exit: Question framed, context available.
Phase 1: Choose the Mode
Entry: Question framed.
| Mode | When | What It Does |
|---|---|---|
| Expert Panel | Multiple domains intersect | Simulate 3-5 relevant expert perspectives |
| Devil's Advocate | You have a preferred option | Systematically attack it |
| What-If Analysis | Uncertain about consequences | Trace each option through scenarios |
| Tradeoff Matrix | Comparing options across criteria | Structured weighted comparison |
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.
- 2mo ago First seen · 231 lines · 81 tokens per session scan A ece2bee01710
think is a skill published in the GitHub repository ondrej-svec/heart-of-gold-toolkit (19 stars, last pushed 1mo ago), licensed MIT. It adds 81 tokens to every session and 2,104 once invoked, about $0.0003 per session on Opus 5.5. 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.
think compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| thinkondrej-svec/heart-of-gold-toolkit | Skill | 19 | 1mo ago | MIT | A | 81 | Claude Code |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | today | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | today | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 314k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |