Borrowing it
Nothing to install: this file belongs to saski/arnesto. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/saski/arnesto/main/.agents/skills/refinement-loop/SKILL.mdgit clone --depth 1 https://github.com/saski/arnestoWrote 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/saski/arnesto/refinement-loop)<a href="https://agentmods.dev/skills/saski/arnesto/refinement-loop"><img src="https://agentmods.dev/badge/skills/saski/arnesto/refinement-loop.svg" alt="Measured on agentmods" 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 | $0.00052 | $0.00626 |
| Opus 5 | $0.00026 | $0.00313 |
| Sonnet 5 | $0.00010 | $0.00125 |
| Haiku 4.5 | $0.00005 | $0.00063 |
Grade A, and why
refinement-loop 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 3d 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.
This is a copy
100% identical to refinement-loop — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Refinement Loop
STARTER_CHARACTER = 🔄
Iterative refinement through file artifacts. Each pass removes one layer of noise, revealing the next.
Setup
Ensure playground/ exists and is in .gitignore. All iteration files go there.
Process
1. Clarify Goal (if the user hasn't defined it for you already earlier in the conversation)
Ask the user:
- What are you refining and what is the goal of refinement?
- Derive a short tag that we'll use as filename for iterating:
{goal}-{subject}(e.g.,gist-nullables,simplify-api,distill-auth-docs)
2. Capture Starting Point
Write original to: playground/{goal}-{subject}-0.md
3. Iterate
Loop:
- Read back the current file (forces fresh perspective)
- Reflect critically: What's missing? What's weak? What could be clearer?
- If improvements found: Write improved version to
playground/{goal}-{subject}-{N+1}.md, then loop again
Before Stopping - Exhaustive Check
When you think you're done, you're probably not. Run through this:
- List everything that could still be improved - even small things (formatting, word choice, structure, clarity)
- Consider what we haven't considered - what angles did we miss? What would someone else notice?
- Try improving in a new direction - not just polishing what's there, but questioning assumptions
- Read as if seeing it for the first time - does it immediately make sense? Is anything unclear?
Only stop when you've gone through this checklist extensively multiple times and genuinely found nothing. There is no "good enough" - someone will use this later and shouldn't waste time on mediocre results.
4. Present Final
Show the user the final version with a brief summary of the refinement journey and number of iterations you used. If deeper issues or questions surfaced, present them to the user as well.
Principles
- Read back forces fresh eyes: Reading from file breaks the "I just wrote this" blindness
- No "good enough": Every detail matters. Formatting, word choice, structure - all of it
- Consider the unconsidered: What angles haven't we explored? What would someone else see? Is there something that the user hasn't even considered, too? (you can surface it as questions at the end)
- Files force iterative improvement: Writing to files prevents "pretend" iteration in conversation. Real iteration gets to much better results.
- Earn the stop: Run the exhaustive check. If you find nothing, you've earned stopping.
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.
- 3d ago First seen · 59 lines · 52 tokens per session scan A 154be86c16f5
refinement-loop is a skill published in the GitHub repository saski/arnesto (5 stars, last pushed today), licensed Unlicense. It adds 52 tokens to every session and 626 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to refinement-loop, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
readonly-review
Run a no-network read-only review practice with two explorer steps and one synthesis step.
repo-harness-assessment
Use when evaluating repository agent-readiness, mapping harness roles, choosing the next smallest improvement, or designing and reconciling agent entrypoints such as AGENTS.md, CLAUDE.md, GEMINI.md, Cursor rules, or GitHub instructions.
work-state-and-delivery
Use when designing or reconciling design docs, task boards, external trackers, execution plans, delivery records, handoffs, review evidence, task-to-change traceability, or repository commit-coupling policy.
repo-contracts-and-boundaries
Use when turning architecture, layering, ownership, dependency direction, schemas, structural metrics, quality thresholds, baselines, allowlists, or generated quality snapshots into repository checks.
runtime-evidence-and-tracing
Use when connecting observed behavior, logs, metrics, request IDs, run IDs, screenshots, traces, external dependency results, or artifacts into a runtime evidence loop.
validation-harness-design
Use when designing repository validation commands, doctor scripts, test matrices, JSON or JUnit outputs, CI gates, smoke checks, or harness command surfaces.