Extract Approach

Extract Approach is a skill for Claude Code from ralfyishere/rules-with-receipts. It costs 112 tokens per session (1,373 once invoked), scanned A, original, MIT.

A skill for recording reusable lessons after solving a difficult or unusual problem. It saves a short learning note in the project so later work can benefit from the same approach.

In plain words
What is it for?
Use it after a non-trivial bug fix, architecture decision, difficult prompt correction, or discovered workflow problem when the lesson is likely to help future work.
Why use it?
Important decisions and fixes can be forgotten when a coding session ends. A brief note preserves patterns, failure modes, and corrections without repeating the whole investigation.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents; mentions Claude Code.

Good fit Use it after a non-trivial bug fix, architecture decision, difficult prompt correction, or discovered workflow problem when the lesson is likely to help future work.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ralfyishere/rules-with-receipts/extract-approach
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 ralfyishere/rules-with-receipts --skill extract-approach
Clone the repo
git clone --depth 1 https://github.com/ralfyishere/rules-with-receipts

Made for: Claude Code.

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 Extract Approach

README.md
[![agentmods](https://agentmods.dev/badge/skills/ralfyishere/rules-with-receipts/extract-approach/github.svg)](https://agentmods.dev/skills/ralfyishere/rules-with-receipts/extract-approach)
Your own site
<a href="https://agentmods.dev/skills/ralfyishere/rules-with-receipts/extract-approach"><img src="https://agentmods.dev/badge/skills/ralfyishere/rules-with-receipts/extract-approach/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 Extract Approach

Your own site · 80×15
<a href="https://agentmods.dev/skills/ralfyishere/rules-with-receipts/extract-approach"><img src="https://agentmods.dev/badge/skills/ralfyishere/rules-with-receipts/extract-approach.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 112 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,373 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 $0.00112 $0.01373
Opus 5 $0.00056 $0.00687
Sonnet 5 $0.00022 $0.00275
Haiku 4.5 $0.00011 $0.00137

Measured 10d ago against content hash 573f1febe1e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

Extract Approach 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 10d 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.

.claude/skills/extract-approach/SKILL.md · 85 lines

How it starts

The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Extract Approach

Purpose

Hard-won knowledge evaporates when the session ends. This skill converts a solved problem into a durable asset: a short note future sessions can read in 30 seconds and apply. It is the persistence half of learning — self-improvement-loop corrects course within a session; this skill writes the reusable residue across sessions.

When to use this skill

After (not during) solving something non-trivial, when at least one of these is true:

  • The solution created a new pattern worth reusing (an approach, a decision rule, a fixture design).
  • The work exposed a recurring failure mode (yours, the model's, or the system's).
  • The outcome changes how future work should be done (a tool quirk discovered, a wrong assumption corrected, a workflow that repeated for the second time).

When NOT to use this skill

  • Trivial or routine tasks — a note nobody will need is clutter, and clutter kills the index.
  • Knowledge the repo already records (code comments, git history, existing docs, existing skills). Don't duplicate; link.
  • Mid-task — finish and verify first. Notes about unverified solutions record guesses.

Operating procedure

  1. Check the bar. Does this meet one of the three criteria above? If not, stop — no note.
  2. Check for an existing note or skill covering the same ground (ls .claude/learnings/, skim titles). If one exists, update it instead of creating a near-duplicate.
  3. Write the note to .claude/learnings/YYYY-MM-DD-short-title.md (kebab-case title, today's date) using _template.md:
# <Short title>
- **Problem:** what was broken/needed, one line.
- **Context:** where this applies (project, tool, situation), one line.
- **What worked:** the approach that solved it, 1-3 lines.
- **What failed / almost failed:** the dead ends or near-misses, 1-2 lines.
- **Decision rule:** "Next time X, do Y" — mechanical enough to follow cold.
- **Verification:** how to confirm the approach worked, one line.
- **Related skills:** [links to skills this touches]
- **Disposition:** learning-note | promote-to-skill | update-skill:<name> | update-snippet

Read the full file on GitHub · 85 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. 10d ago First seen · 85 lines · 0 tokens per session scan A 573f1febe1e8

Subscribe to this mod's changes

Extract Approach is a skill published in the GitHub repository ralfyishere/rules-with-receipts (2 stars, last pushed 2mo ago), licensed MIT. It adds 112 tokens to every session and 1,373 once invoked, about $0.0006 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