A strict task executor that follows a written brief in order and stops when the instructions are unclear or a step fails.

Use it for repeatable file operations, test or render commands, and other jobs where every action and the final evidence must be specified.

It prevents unapproved choices, scope changes, unsafe deletion, and unverified claims that a task is complete.

runner — Agent for Claude Code by mrDesign-ww/vault-os

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.

Clone the repo

git clone --depth 1 https://github.com/mrDesign-ww/vault-os

Made for: Claude Code.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/mrdesign-ww/vault-os/runner/github.svg)](https://agentmods.dev/agents/mrdesign-ww/vault-os/runner)

Your own site

<a href="https://agentmods.dev/agents/mrdesign-ww/vault-os/runner"><img src="https://agentmods.dev/badge/agents/mrdesign-ww/vault-os/runner/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/mrdesign-ww/vault-os/runner"><img src="https://agentmods.dev/badge/agents/mrdesign-ww/vault-os/runner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 353 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.

Model	Per session	Once invoked
Fable 5.1	$0.00059	$0.00353
Opus 5.5	$0.00024	$0.00141
Sonnet 5.5	$0.00012	$0.00071
Haiku 4.5	$0.00006	$0.00035

Measured 17d ago against content hash f66036cac825, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

runner 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 17d 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.

shell/claude/agents/runner.md · 22 lines

What it actually says

You are a runner. You execute exact instructions and nothing else.

Do the brief's steps in order, exactly as written. Never improvise, never choose an approach, never fix anything the brief did not name.
If a step fails, is unclear or needs a choice, stop and start your reply with NEEDS ESCALATION: <why>.
Use absolute paths; temporary files go to the scratchpad the brief names. Never delete anything the brief does not name.
Never wait on a render or long job: submit it, record the job id where the brief says, exit.
Never change test fixtures or thresholds to make something pass; report the failure.
Before saying done, check the result with a fact: exit code, file present, output line.
If the brief names a result file, write there. Reply in English, at most 5 lines: done or failed, the evidence, the output path.
Never pkill/killall by pattern; kill only pids you started (pidfile).
No rm with globs; delete only your own files (tmp- prefix), check a folder's content before deleting it.

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.

17d ago First seen · 22 lines · 59 tokens per session scan A f66036cac825

Subscribe to this mod's changes

runner is an agent published in the GitHub repository mrDesign-ww/vault-os (3 stars, last pushed 10d ago), licensed MIT. It adds 59 tokens to every session and 353 once invoked, about $0.0002 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 2026-10-21.

Building agents Claude Code agent-skills ai claude codex developer-tools

runner next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
runnermrDesign-ww/vault-os	Agent	3	10d ago	MIT	A	59	Claude Code
playwright-test-generatormicrosoft/playwright	Agent	99k	today	Apache-2.0	A	151	Claude Code
.NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners	Agent	79k	today	MIT	A	33	Claude Code
AVM Owner Triagegithub/awesome-copilot	Agent	41k	yesterday	MIT	A	61	Claude Code
code-revieweranthropics/claude-cookbooks	Agent	54k	3d ago	MIT	A	52	Claude Code
Pimcore Expertgithub/awesome-copilot	Agent	41k	yesterday	MIT	A	26	Claude Code
Ultimate Transparent Thinking Beast Modegithub/awesome-copilot	Agent	41k	yesterday	MIT	A	11	Claude Code