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 somnus0x/agt-skill-pack --skill hold-upgit clone --depth 1 https://github.com/somnus0x/agt-skill-packWrote 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/somnus0x/agt-skill-pack/hold-up)<a href="https://agentmods.dev/skills/somnus0x/agt-skill-pack/hold-up"><img src="https://agentmods.dev/badge/skills/somnus0x/agt-skill-pack/hold-up/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/somnus0x/agt-skill-pack/hold-up"><img src="https://agentmods.dev/badge/skills/somnus0x/agt-skill-pack/hold-up.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 | $0.00097 | $0.01491 |
| Opus 5.5 | $0.00039 | $0.00596 |
| Sonnet 5.5 | $0.00019 | $0.00298 |
| Haiku 4.5 | $0.00010 | $0.00149 |
Grade A, and why
hold-up 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 1mo 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hold Up — Clarify Before You Act
You ask "should we do X?" and the AI has already done X. You think out loud — "i think maybe we move this to a queue" — and it's three files deep into building a queue. You wanted to talk; it heard an order.
This is the #1 friction of agentic coding: the AI reads exploratory messages as commands, and acts on inferred scope. Hold Up is the brake. It catches the AI mid-sprint, before the wrong action, and trades one clarifying turn for the rework-and-trust cost of building the wrong thing.
Why
AI coding tools are biased toward action. Ambiguity gets resolved by doing, not asking — because doing looks helpful. But a question costs one turn; a wrong action costs the rework, the cleanup, and a chunk of your trust that it understood you. The asymmetry is brutal and one-directional.
The failure isn't "the AI is dumb." It's that it can't tell discuss-mode from do-mode, so it defaults to do. Hold Up installs the missing check: is this person asking me to act, or thinking out loud? — run before any non-trivial action, not after.
How it differs from grill-me and PLAN mode
grill-me= a deliberate, invoked, exhaustive interview. You summon it to walk the entire design tree before a big build. Heavyweight, on-demand.- PLAN mode = "design the implementation, don't edit yet." Still assumes you've decided to build.
- Hold Up = automatic, reflexive, one question. It fires itself when it smells discuss-mode, asks the single thing it needs, and either releases or escalates. Lightweight, always-on.
The chain: Hold Up catches "wait, do you even want me to build?" → if yes and it's big → escalate to grill-me → then build. Hold Up is the front door; grill-me is the interrogation room.
When to use
Explicit
Say /hold-up, "just talk", "don't action this", "let's think first", "discuss before you build", "wait wait".
Auto-fire (the real value — add to your CLAUDE.md)
Fire Hold Up automatically when both are true:
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.
- 1mo ago First seen · 105 lines · 97 tokens per session scan A dbdf1bacc890
hold-up is a skill published in the GitHub repository somnus0x/agt-skill-pack (74 stars, last pushed 18d ago), licensed Apache-2.0. It adds 97 tokens to every session and 1,491 once invoked, about $0.0004 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.
Other skills, from other repositories
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…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…