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 ondrej-svec/heart-of-gold-toolkit --skill groundgit clone --depth 1 https://github.com/ondrej-svec/heart-of-gold-toolkitWrote 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/ground)<a href="https://agentmods.dev/skills/ondrej-svec/heart-of-gold-toolkit/ground"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/ground/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/ground"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/ground.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00116 | $0.03226 |
| Opus 5 | $0.00058 | $0.01613 |
| Sonnet 5 | $0.00023 | $0.00645 |
| Haiku 4.5 | $0.00012 | $0.00323 |
Grade A, and why
ground 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 12d 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 — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ground
Survey the current repo, then ground externally on what's relevant. Dynamic per-repo — no static config, no curated tech list. The repo IS the configuration.
Boundaries
This skill MAY: read manifests, scan git history, fetch from the web (WebSearch/WebFetch/context7/gh), write the briefing file at docs/ground/<fingerprint>.md.
This skill MAY NOT: edit code, modify configuration, install dependencies, run anything other than read-only gh queries. Briefings are advisory artifacts.
Designed to run as a subagent. Web research bloats context fast — keep the heavy fetching in an isolated context and return only the synthesized briefing to the caller.
Common Rationalizations
| Shortcut | Why It Fails | The Cost |
|---|---|---|
| "Skip the local survey — just web search for the topic" | The repo's stack disambiguates the search. Without it, results are generic. | Wallpaper briefings that don't change Claude's behavior |
| "Fetch everything in parallel, ignore timeouts" | One slow source blocks the briefing. Worse: stalls the calling skill. | /work latency balloons; Ondrej learns to skip /ground |
| "Synthesize without citing sources" | Looks confident, can't be verified, indistinguishable from training-data regurgitation | Defeats the entire point — briefings must be fact-checkable |
| "Always refresh — never trust the cache" | Re-fetches every time, makes cross-session reuse impossible | Token/time cost on every invocation; cache pattern wasted |
| "Generate a long briefing — more detail is better" | Past 1500 tokens the briefing gets ignored as wallpaper | Token waste + behavioral degradation |
Phase 0: Detect Inputs
Entry: Skill invoked standalone (slash command or Task call from another skill).
Resolve:
cwd— current working directory. From the invocation context.topic— what the user (or calling skill) is grounding for. From the invocation arguments. Required; if absent, ask via AskUserQuestion.local_only— boolean flag. Defaultfalse. Whentrue, skip Phase 4 (external grounding).force_refresh— boolean flag. Defaultfalse. Whentrue, skip cache check in Phase 3.invocation_mode—standaloneif a user invoked/grounddirectly (slash command);embeddedif another skill or agent handed you this protocol via a Task/subagent call.
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.
- 12d ago First seen · 235 lines · 116 tokens per session scan A c6d9bf58f2d5
ground is a skill published in the GitHub repository ondrej-svec/heart-of-gold-toolkit (19 stars, last pushed 22d ago), licensed MIT. It adds 116 tokens to every session and 3,226 once invoked, about $0.0006 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…