hold-my-hand

hold-my-hand is a skill for Claude Code from melodic-software/claude-code-plugins. It costs 130 tokens per session (1,947 once invoked), scanned A, original, MIT.

A guided-session skill that breaks a long job into phases and shows only the current phase. Each phase ends at a point the user can check, such as a file being created or a service responding.

In plain words
What is it for?
Use it when you want step-by-step guidance with exact commands, expected results, common failures, and confirmation points.
Why use it?
It prevents a large set of instructions from arriving all at once and gives the user a clear checkpoint before continuing.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter.

Part of the discipline plugin — 18 skills shipped together

Good fit Use it when you want step-by-step guidance with exact commands, expected results, common failures, and confirmation points.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/melodic-software/claude-code-plugins/hold-my-hand
Install

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.

Any agent
npx skills add melodic-software/claude-code-plugins --skill hold-my-hand
Clone the repo
git clone --depth 1 https://github.com/melodic-software/claude-code-plugins

Made for: Claude Code.

Or install discipline, the plugin that ships this one along with the rest of its 18 skills.

Wrote 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.

agentmods badge for hold-my-hand

README.md
[![agentmods](https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/hold-my-hand/github.svg)](https://agentmods.dev/skills/melodic-software/claude-code-plugins/hold-my-hand)
Your own site
<a href="https://agentmods.dev/skills/melodic-software/claude-code-plugins/hold-my-hand"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/hold-my-hand/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.

agentmods 80×15 button for hold-my-hand

Your own site · 80×15
<a href="https://agentmods.dev/skills/melodic-software/claude-code-plugins/hold-my-hand"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/hold-my-hand.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 130 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,947 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00130 $0.01947
Opus 5.5 $0.00052 $0.00779
Sonnet 5.5 $0.00026 $0.00389
Haiku 4.5 $0.00013 $0.00195

Measured 11d ago against content hash 1c5843d923d7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

hold-my-hand 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 11d 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.

plugins/discipline/skills/hold-my-hand/SKILL.md · 168 lines

How it starts

The opening of the file, as written. The whole thing — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hold my hand

Once invoked, this is a standing instruction for the rest of the session. Take the work that remains, cut it into phases, and present one phase at a time. The user never sees Phase 2 until Phase 1 is presented and done.

A declared species in this plugin: NOT a corrector

This skill is a standing session posture over how future work is delivered, not a re-anchor/audit/correct loop. It audits nothing; it changes the shape of every subsequent response until the session ends or the user cancels it. It carries no discipline-batch tier, on purpose: a batch sweep must never impose a delivery posture the user did not ask for.

Partition the work into phases

A phase is a contiguous run of steps that ends at a point the user can confirm. Cut on the confirmable boundary, not on an even step count: a phase ends where something is observably true (a file exists, a service answers, a page loads).

State the count once, on a single line, so the user knows the shape of the job without reading ahead:

Phase 3 of 8: Install the runner on the build machine

Never list the later phases. Not as a preview, not as a table of contents, not as a "here is where we are going" paragraph, and not as a task-ledger dump the user reads over your shoulder. The count is the only forward information.

When the work turns out to need more phases than the first estimate, say so in one line ("this is now 10 phases, not 8") and continue. Never grow the count silently.

What every phase states

Each phase carries all six of these. A phase missing one of them is not ready to present.

  1. Where. The machine, the kind of terminal (PowerShell, bash, a browser tab, a physical console), whether it must be elevated, and the working directory.
  2. Numbered steps. One action per numbered step, in the order they run.
  3. The exact command, in a fenced block, with real values substituted, never a template with placeholders the user must fill in from memory. A step that has no command because it happens in a browser, a vendor dashboard, or on a physical console names the exact control to operate and where to find it, in the command's place. Never invent a command to satisfy this part.
  4. The expected output or observable result after each command or action, so the user can tell success from failure without asking.
  5. The common failure and its repair. The one that actually happens, with what to do about it.
  6. The exact reply that advances. Write it as a literal the user can copy:

Read the full file on GitHub · 168 lines

Files

What ships with it

1 file 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.

Changes

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.

  1. 11d ago Changed · -6 lines 1c5843d923d7
  2. 28d ago Changed · -87 tokens per session 35e6a391a266
  3. 1mo ago First seen · 174 lines · 217 tokens per session scan A de2032bf06bf

Subscribe to this mod's changes

hold-my-hand is a skill published in the GitHub repository melodic-software/claude-code-plugins (28 stars, last pushed today), licensed MIT. It adds 130 tokens to every session and 1,947 once invoked, about $0.0005 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 2026-09-07.