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 GregoryHo/arcforge --skill loopinggit clone --depth 1 https://github.com/GregoryHo/arcforgeWrote 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/gregoryho/arcforge/looping)<a href="https://agentmods.dev/skills/gregoryho/arcforge/looping"><img src="https://agentmods.dev/badge/skills/gregoryho/arcforge/looping/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/gregoryho/arcforge/looping"><img src="https://agentmods.dev/badge/skills/gregoryho/arcforge/looping.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 | $0.00018 | $0.01920 |
| Opus 5.5 | $0.00007 | $0.00768 |
| Sonnet 5.5 | $0.00004 | $0.00384 |
| Haiku 4.5 | $0.00002 | $0.00192 |
Grade A, and why
looping 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 today.
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 — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Looping
One task per iteration, each in its own fresh session. Nothing carries over
between them except files: the task list holds what is left, git holds the work,
and .arcforge-loop.json holds the loop's own bookkeeping. A session can die, a
machine can sleep, and the next iteration still knows where it was — and nobody
is watching. Every limit you want has to be a flag typed before you walk away.
Step 1 — The list exists before the loop does
The loop's only task source is a markdown checkbox task list passed with
--tasks. There is no second task format and no implicit default file.
> arcforge task list v1
- [ ] T1 — Extract the tokenizer into its own module
- verify: `npm test`
- [ ] T2 — Wire the tokenizer into the reader
- verify: `npm test`
Each task is one unattended session's worth of work, written so a reader with no
context can act on it. Its verify: line is that task's acceptance floor: the
loop runs it after the session exits and refuses to mark the task done when it
fails. A task without one falls back to the run-wide --verify-cmd; a task with
neither is accepted on the session's exit code alone, which only says the session
ended, not that the work happened. Run the project's own suite before launching
too: the loop cannot tell a baseline that was already broken from a task that
broke it, so it will spend a session, fail the floor, retry, and block a task that
was never the problem.
- Done when the list exists and parses, every task carries a
verify:line or the run supplies--verify-cmd, and the suite has been run here and is green.
Step 2 — Fix the stop conditions before launching
Three ceilings are yours to set — --max-runs bounds iterations (default 50),
--max-cost bounds dollars across the whole run (default: no ceiling), and
--task-timeout bounds one session's seconds (default 600). The rest fire on their
own. "Let it get as far as it can overnight" is the request to decline, not the
instruction to follow: guessing a ceiling too low costs one more run, while having
no ceiling is unbounded by definition and the loop is spending money on a
repository nobody is reading.
What ships with it
2 files 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.
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.
- today First seen · 156 lines · 18 tokens per session scan A b41aeff018cf
looping is a skill published in the GitHub repository GregoryHo/arcforge (8 stars, last pushed 12d ago), licensed MIT. It adds 18 tokens to every session and 1,920 once invoked, about $0.0001 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 2026-11-07.
looping compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| loopingGregoryHo/arcforge | Skill | 8 | 12d ago | MIT | A | 18 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |