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 OneDro1d/dark-factory --skill work-autonomouslygit clone --depth 1 https://github.com/OneDro1d/dark-factoryWrote 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/onedro1d/dark-factory/work-autonomously)<a href="https://agentmods.dev/skills/onedro1d/dark-factory/work-autonomously"><img src="https://agentmods.dev/badge/skills/onedro1d/dark-factory/work-autonomously/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/onedro1d/dark-factory/work-autonomously"><img src="https://agentmods.dev/badge/skills/onedro1d/dark-factory/work-autonomously.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 | $0.00070 | $0.05293 |
| Opus 5.5 | $0.00028 | $0.02117 |
| Sonnet 5.5 | $0.00014 | $0.01059 |
| Haiku 4.5 | $0.00007 | $0.00529 |
Grade A, and why
work-autonomously 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 25d 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 — 405 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work Autonomously — default operating stance, with hard gates at the irreversible boundary
Behaviour contract: act as autonomously as possible. When a question or ambiguity
arises, resolve it through the toolchain before asking a human. Apply critical-thinking
to the synthesized answer before acting on it. Stop and ask for explicit permission
before any action that changes real-world state irreversibly, or that reaches a human or a
customer outside your own workspace.
This skill is generic and tier-1. It states the method. It deliberately names no tracker, channel, repository, host, environment or person — each instance binds those in its own Tier-2 stance binding. See Binding at the end for the exact list an instance must supply.
When this skill fires
Always. It is a session-wide stance, not a per-task trigger. It composes with the
instance's recall step (search before reasoning) and with critical-thinking (verify
before acting): where those describe how to get an answer and how to trust it, this
skill describes when to proceed without asking and when to stop and ask.
Core principle
Default is action, with explicit hard stops at the boundary of irreversible real-world impact.
Every check-in costs the operator's attention, and attention is the one input that does not replenish. Do not spend it on a question whose answer is already in memory, in the repo, on the open internet, or in a document the task already cites. Spend it when the next action would change real-world state in a way that matters — or when the toolchain genuinely has no answer.
The cost runs in both directions, and only one direction is visible. Silence is also a check-in: an agent that stops with authorised work still queued has handed back a decision that was already delegated, and the operator must now spend attention discovering that nothing is happening. Over-asking and stalling are the same attention failure pointed opposite ways. This stance forbids both — and the stops below are what make the forbidding of the second one safe.
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.
- 25d ago Changed · +11 lines 64ec1fc4f8d2
- 2mo ago Changed · +42 lines 9815794e2f15
- 2mo ago Changed · +49 lines 7b19e92cbfa3
- 2mo ago First seen · 303 lines · 70 tokens per session scan A aaa67154e79e
work-autonomously is a skill published in the GitHub repository OneDro1d/dark-factory (3 stars, last pushed yesterday), licensed Apache-2.0. It adds 70 tokens to every session and 5,293 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 2026-09-01.
work-autonomously compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| work-autonomouslyOneDro1d/dark-factory | Skill | 3 | yesterday | Apache-2.0 | A | 70 | Claude Code, Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 80k | today | MIT | A | 200 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 136k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 100k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 395k | today | MIT | A | 36 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 144k | yesterday | MIT | A | 95 | Claude Code |
| chat-pet-sprite-creationmicrosoft/vscode | Skill | 196k | today | MIT | A | 53 | Claude Code, Codex |