bead-polishing-ralph

bead-polishing-ralph is a skill for Claude Code, Codex from vnnkl/agentflywheel. It costs 140 tokens per session (4,322 once invoked), scanned A, original, MIT.

A procedure for refining beads, which are small units of work for the ralph-tui task runner. It checks that each bead fits in one agent session, has quality checks, uses the correct command syntax, and follows task dependencies.

In plain words
What is it for?
Use it to inspect a plan or product requirements document, split work into executable beads, polish their descriptions, attach quality gates, and order database, backend, and user-interface tasks correctly.
Why use it?
It reduces failures caused by tasks that are too large, vague, incorrectly ordered, or missing validation instructions. This matters because ralph-tui gives each bead to a fresh agent without earlier context.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to inspect a plan or product requirements document, split work into executable beads, polish their descriptions, attach quality gates, and order database, backend, and user-interface tasks correctly.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/vnnkl/agentflywheel/bead-polishing-ralph
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.

Any agent
npx skills add vnnkl/agentflywheel --skill bead-polishing-ralph
Clone the repo
git clone --depth 1 https://github.com/vnnkl/agentflywheel

Made for: Claude Code, Codex.

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.

agentmods badge for bead-polishing-ralph

README.md
[![agentmods](https://agentmods.dev/badge/skills/vnnkl/agentflywheel/bead-polishing-ralph/github.svg)](https://agentmods.dev/skills/vnnkl/agentflywheel/bead-polishing-ralph)
Your own site
<a href="https://agentmods.dev/skills/vnnkl/agentflywheel/bead-polishing-ralph"><img src="https://agentmods.dev/badge/skills/vnnkl/agentflywheel/bead-polishing-ralph/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.

agentmods 80×15 button for bead-polishing-ralph

Your own site · 80×15
<a href="https://agentmods.dev/skills/vnnkl/agentflywheel/bead-polishing-ralph"><img src="https://agentmods.dev/badge/skills/vnnkl/agentflywheel/bead-polishing-ralph.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 140 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,322 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00140 $0.04322
Opus 5 $0.00070 $0.02161
Sonnet 5 $0.00028 $0.00864
Haiku 4.5 $0.00014 $0.00432

Measured 9d ago against content hash a7a21b0cc53a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

bead-polishing-ralph 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 9d 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.

skills/bead-polishing-ralph/SKILL.md · 353 lines

How it starts

The opening of the file, as written. The whole thing — 353 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bead Polishing for Ralph-TUI

Iteratively refine beads through 4-6 polishing rounds until convergence, enforcing all constraints that ralph-tui needs for successful autonomous execution.

This skill combines the convergence-tracked polishing loop from the Agentic Coding Flywheel with the specific format and sizing constraints required by ralph-tui's beads tracker.

Why This Matters

ralph-tui spawns a fresh agent instance per bead with no memory of previous work. A bead that is too big, vague, or has broken dependencies will cause the agent to either run out of context, produce the wrong thing, or block the entire pipeline. Polishing beads before execution is the single highest-ROI activity because:

  • Plan mistakes cost 1x to fix (edit the markdown)
  • Bead mistakes cost 5x to fix (rewrite + re-link + re-order)
  • Code mistakes cost 25x to fix (debug + refactor + retest across files)

A 30-second fix to a bead description prevents a 30-minute failed ralph-tui iteration.


Before Starting

Detect CLI

Use bd (beads) as the CLI for all bead operations. The polishing process uses bd commands throughout.

Gather Context

  1. Read the source plan/PRD that the beads were created from. Polishing without the source is guessing.
  2. Extract Quality Gates from the PRD's "Quality Gates" section. If none exists, ask the user what commands should pass (e.g., pnpm typecheck, pnpm lint).
  3. Check if beads exist: bd list --json

Create Beads if None Exist

If no beads exist yet, create them from the source PRD before polishing. This makes the separate "convert to tasks" step optional — bead-polishing-ralph can handle the full plan-to-polished-beads pipeline.

  1. Create an epic linking back to the PRD:

    bd create --type=epic \
      --title="[Feature Name]" \
      --description="$(cat <<'EOF'
    [Feature description from PRD]
    EOF
    )" \
      --external-ref="prd:./path-to-prd.md"
    
  2. Create one bead per user story/requirement from the PRD. For each story:

    • Title matches the story title
    • Description includes story context, acceptance criteria, quality gates, and the <promise>COMPLETE</promise> instruction
    • Priority reflects dependency order (schema=1, backend=2, UI=3, polish=4)
    • Pass --parent=<epic-id> at creation time so the parent link exists from the start

Read the full file on GitHub · 353 lines

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.

  1. 9d ago First seen · 353 lines · 140 tokens per session scan A a7a21b0cc53a

Subscribe to this mod's changes

bead-polishing-ralph is a skill published in the GitHub repository vnnkl/agentflywheel (2 stars, last pushed 3mo ago), licensed MIT. It adds 140 tokens to every session and 4,322 once invoked, about $0.0007 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 2026-08-31.

Related

Other skills, from other repositories