learn-mistake

learn-mistake is a command for Claude Code from liortesta/ClawdAgent. It costs 0 tokens per session (335 once invoked), scanned A, original, Apache-2.0.

A self-correction procedure for recording mistakes made during development. It explains the cause, creates a specific prevention rule, and saves it in CLAUDE.md, a project instruction file.

In plain words
What is it for?
Use it after an agent makes a mistake, such as using the wrong package manager, putting a file in the wrong directory, or skipping tests.
Why use it?
It turns repeated mistakes into written project rules instead of relying on memory. It can also prompt you to check whether an automated hook could catch the same problem.

Command for Claude Code

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.

agentmods
npx agentmods add commands/liortesta/clawdagent/learn-mistake
Clone the repo
git clone --depth 1 https://github.com/liortesta/ClawdAgent

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 learn-mistake

README.md
[![agentmods](https://agentmods.dev/badge/commands/liortesta/clawdagent/learn-mistake.svg)](https://agentmods.dev/commands/liortesta/clawdagent/learn-mistake)
Your own site
<a href="https://agentmods.dev/commands/liortesta/clawdagent/learn-mistake"><img src="https://agentmods.dev/badge/commands/liortesta/clawdagent/learn-mistake.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 335 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00000 $0.00335
Opus 5 $0.00000 $0.00168
Sonnet 5 $0.00000 $0.00067
Haiku 4.5 $0.00000 $0.00034

Measured 4d ago against content hash b93a15e1d0a5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

learn-mistake 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 4d 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/commands/learn-mistake.md · 29 lines

What it actually says

Learn From Mistake

The user is telling you about a mistake you just made. This is a learning opportunity.

Process:

  1. Acknowledge the specific mistake — be precise about what went wrong
  2. Explain WHY it was wrong — root cause, not just symptoms
  3. Propose a rule to prevent recurrence — specific and actionable
  4. Add the rule to CLAUDE.md under "Self-Correction Rules (LEARNED)" section with today's date
  5. Check if a hook can enforce it — if the mistake is about code patterns, consider adding a PostToolUse hook that catches it automatically

Rule Format in CLAUDE.md:

- [DATE] RULE: [specific, actionable rule]
  CONTEXT: [what happened that triggered this rule]

Examples:

  • [2025-01-15] RULE: Always use pnpm, never npm, in this project. CONTEXT: Used npm install which created package-lock.json conflict.
  • [2025-01-16] RULE: API routes go in src/app/api/, not src/routes/. CONTEXT: Created route in wrong directory, 404 error.
  • [2025-01-17] RULE: Always run tests before committing, even for "trivial" changes. CONTEXT: One-line fix broke 3 tests.

Rules:

  • NEVER be defensive — own the mistake
  • ALWAYS make the rule specific enough to be actionable
  • ALWAYS verify the rule is added to CLAUDE.md before completing

Mistake: $ARGUMENTS

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. 4d ago First seen · 29 lines · 0 tokens per session scan A b93a15e1d0a5

Subscribe to this mod's changes

learn-mistake is a command published in the GitHub repository liortesta/ClawdAgent (11 stars, last pushed 7d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 335 tokens. 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.