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 rornic/zweig --skill zweig-refinegit clone --depth 1 https://github.com/rornic/zweigWrote 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/rornic/zweig/zweig-refine)<a href="https://agentmods.dev/skills/rornic/zweig/zweig-refine"><img src="https://agentmods.dev/badge/skills/rornic/zweig/zweig-refine/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/rornic/zweig/zweig-refine"><img src="https://agentmods.dev/badge/skills/rornic/zweig/zweig-refine.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00079 | $0.02071 |
| Opus 5 | $0.00039 | $0.01035 |
| Sonnet 5 | $0.00016 | $0.00414 |
| Haiku 4.5 | $0.00008 | $0.00207 |
Grade A, and why
zweig-refine 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 12d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Zweig Refine
Refinement is a loop of real edits against a real file, not a pass reasoned about silently and never checked. See EXAMPLE.md for a worked transcript of the loop end to end.
Set up the working draft
- If the draft already lives in a real file you're editing in place (a docs page, a README, a code comment in situ), work directly on that file. Skip straight to the loop.
- Otherwise (a commit message, PR description, Slack message, or any output with no backing file), write the current draft to
<scratchpad>/zweig-draft.md, resolving your session's actual scratchpad path. Do every cut as an Edit against that file, and copy the converged text into the real output only once the loop and gate below are both done. - If the draft is only part of a larger file (one comment in a source file, one section of a template), scope every edit to that region. Unrelated surrounding content doesn't count toward convergence either way.
The loop
- Read the working file in full.
- Cut and rewrite: make an Edit call for every sentence, bullet, or word that can be removed, reworded, or rewritten without losing precise meaning. Three moves count here: cutting words that add nothing, rewording to say the same thing in fewer or plainer words, and rewriting a sentence outright when the same point lands better in a different construction (a stronger verb, an active subject, a reordered clause), even when the result isn't shorter. Throw ballast overboard, favoring fewer, denser points over breadth, and cut self-descriptive or boilerplate lines. In the same pass, fix any em dash, stray semicolon, hedge word, throat-clearing opener, or boilerplate transition, whether it was already there or one of your own edits just introduced it (see "Skip on the way in" and "Punctuation" below). See "Cut slack, not substance" for what's off-limits, and "Rewrite, don't pad" for the line between a full rewrite and scope creep. A cut, reword, or rewrite only counts if it's a real Edit call, not reasoning about one without making it.
- If step 2 made at least one edit and you're under 4 passes total, go back to step 1 for a new pass on the updated file. Otherwise, stop and move to the structural variance pass below.
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.
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.
- 12d ago First seen · 89 lines · 79 tokens per session scan A 8d654c1609fd
zweig-refine is a skill published in the GitHub repository rornic/zweig (11 stars, last pushed 20d ago), licensed MIT. It adds 79 tokens to every session and 2,071 once invoked, about $0.0004 per session on Opus 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
ai-engineering-toolkit
6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.
promptfoo-provider-setup
Configure promptfoo providers or redteam targets for hosted models, live HTTP APIs, Python/JavaScript local scripts, agent SDKs, or multi-input systems. Use when connecting promptfoo to the system under test, mapping vars, auth env vars, request bodies, response transforms, or static-code-derived provider wrappers. Do…
avoid-ai-writing
Audit and rewrite content to remove AI writing patterns ("AI-isms"). Use this skill when asked to "remove AI-isms," "clean up AI writing," "edit writing for AI patterns," "audit writing for AI tells," or "make this sound less like AI." Supports a detect-only mode, an edit-in-place mode for files, an optional voice…
unlazy
Enforces completion discipline for substantial autonomous work by writing acceptance gates before execution, decomposing work with the Depth Tree, running approved checks, and re-verifying evidence before reporting. Use when an agent faces a long or multi-part task, work that has returned half-done, an exhaustive…
prompt-master
Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other…
avoid-ai-writing-router
Use when a request combines AI-writing audit, rewrite, file editing, voice preservation, false-positive interpretation, verification, or when the user invokes Avoid AI Writing without naming a mode.