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.
git clone --depth 1 https://github.com/naimkatiman/continuous-improvementWrote 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/commands/naimkatiman/continuous-improvement/learn-eval)<a href="https://agentmods.dev/commands/naimkatiman/continuous-improvement/learn-eval"><img src="https://agentmods.dev/badge/commands/naimkatiman/continuous-improvement/learn-eval/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/commands/naimkatiman/continuous-improvement/learn-eval"><img src="https://agentmods.dev/badge/commands/naimkatiman/continuous-improvement/learn-eval.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00029 | $0.01196 |
| Opus 5.5 | $0.00012 | $0.00478 |
| Sonnet 5.5 | $0.00006 | $0.00239 |
| Haiku 4.5 | $0.00003 | $0.00120 |
Grade A, and why
learn-eval 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 1mo 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
97% identical to learn-eval — 8 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/learn-eval - Extract, Evaluate, then Save
Extends /learn with a quality gate, save-location decision, and knowledge-placement awareness before writing any skill file.
What to Extract
Look for:
- Error Resolution Patterns — root cause + fix + reusability
- Debugging Techniques — non-obvious steps, tool combinations
- Workarounds — library quirks, API limitations, version-specific fixes
- Project-Specific Patterns — conventions, architecture decisions, integration patterns
Process
-
Review the session for extractable patterns
-
Identify the most valuable/reusable insight
-
Determine save location:
- Ask: "Would this pattern be useful in a different project?"
- Global (
~/.claude/skills/learned/): Generic patterns usable across 2+ projects (bash compatibility, LLM API behavior, debugging techniques, etc.) - Project (
.claude/skills/learned/in current project): Project-specific knowledge (quirks of a particular config file, project-specific architecture decisions, etc.) - When in doubt, choose Global (moving Global → Project is easier than the reverse)
-
Draft the skill file using this format:
---
name: pattern-name
description: "Under 130 characters"
user-invocable: false
origin: auto-extracted
---
# [Descriptive Pattern Name]
**Extracted:** [Date]
**Context:** [Brief description of when this applies]
## Problem
[What problem this solves - be specific]
## Solution
[The pattern/technique/workaround - with code examples]
## When to Use
[Trigger conditions]
-
Quality gate — Checklist + Holistic verdict
5a. Required checklist (verify by actually reading files)
Execute all of the following before evaluating the draft:
- Grep
~/.claude/skills/and relevant project.claude/skills/files by keyword to check for content overlap - Check MEMORY.md (both project and global) for overlap
- Consider whether appending to an existing skill would suffice
- Confirm this is a reusable pattern, not a one-off fix
- Grep
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.
- 1mo ago First seen · 118 lines · 29 tokens per session scan A 62284005edcd
learn-eval is a command published in the GitHub repository naimkatiman/continuous-improvement (9 stars, last pushed 1mo ago), licensed MIT. It adds 29 tokens to every session and 1,196 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 97% identical to learn-eval, differing in 8 lines, and is treated as a copy.
Other commands, from other repositories
reflect-promote
Promote learnings from skills to global CLAUDE.md.
rem_help
Each section is printed verbatim by --help for the command it names. The agreed design is issue #1656; change a page there and here together.
compact-prep
Ask the agent to prepare for conversation compaction by updating any relevant state and providing guidance for the compaction agent and to kick off the session there after.
speckit.archive
Archive a feature specification into main project memory after merge, resolving gaps and conflicts.
fire-reflect
After any failure (debug resolution, test failure, approach rotation, stalled loop), capture what was tried, why it failed, and what actually worked as a persistent reflection. Future sessions search these before investigating.
sync
Mid-session context refresh command. Keeps context fresh and consistent during work.