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.
npx agentmods add skills/rtazima/claude-proj-blueprint/learnernpx skills add rtazima/claude-proj-blueprint --skill learnergit clone --depth 1 https://github.com/rtazima/claude-proj-blueprintWrote 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/rtazima/claude-proj-blueprint/learner)<a href="https://agentmods.dev/skills/rtazima/claude-proj-blueprint/learner"><img src="https://agentmods.dev/badge/skills/rtazima/claude-proj-blueprint/learner.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 | $0.00051 | $0.01396 |
| Opus 5 | $0.00026 | $0.00698 |
| Sonnet 5 | $0.00010 | $0.00279 |
| Haiku 4.5 | $0.00005 | $0.00140 |
Grade A, and why
learner 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Learner
Analyze completed work and extract patterns to improve existing skills, suggest new ones, and refine project conventions. This is a meta-skill: it makes other skills better.
Rules
- Read-only analysis first — understand patterns before suggesting changes
- Evidence-based — every suggestion must reference specific commits, files, or patterns
- Conservative — suggest changes, don't auto-apply. The human decides what to adopt
- Quality gate — only suggest patterns that appear 3+ times (not one-offs)
- Output to docs — write the report to
docs/architecture/learner-report-{date}.md
Workflow
Phase 1: Gather data
- Run
git log --oneline -N(N from args, default 20) to see recent commits - Run
git log --stat -Nto see which files changed most - Read the diffs of significant commits:
git show --stat <sha> - Read existing skills in
.claude/skills/*/SKILL.md - Read existing commands in
.claude/commands/*.md - Read
CLAUDE.mdfor current conventions
Phase 2: Analyze patterns
Look for:
A. Skill gaps
- Recurring manual steps that no skill covers
- Patterns Claude keeps explaining/doing that could be codified
- Workarounds or repeated code patterns across files
B. Skill improvements
- Skill rules that were violated in practice (maybe they're wrong)
- Skill checklists that are consistently skipped (maybe too strict)
- Skill triggers that didn't fire when they should have (description needs updating)
- Missing examples in skills that would have helped
C. Convention drift
- CLAUDE.md rules that don't match actual practice
- Commit message style drift from Conventional Commits
- File naming patterns that evolved beyond what CLAUDE.md documents
D. Hook gaps
- Errors that hooks should have caught but didn't
- Repetitive manual checks that could be automated
- New file types/patterns not covered by existing hooks
Phase 3: Report
Write a structured report:
# Learner Report — {date}
## Analyzed
- Commits: {N} (from {oldest_sha} to {newest_sha})
- Files changed: {count}
- Period: {date_range}
## Suggested new skills
### 1. [skill-name]
- **Pattern observed**: [what was done manually 3+ times]
- **Evidence**: commits {sha1}, {sha2}, {sha3}
- **Proposed trigger**: "[keywords]"
- **Proposed rules**: [list]
## Suggested skill improvements
### 1. [existing-skill-name]
- **Current rule**: [what the skill says]
- **Observed practice**: [what actually happened]
- **Suggestion**: [specific change]
- **Evidence**: [commits/files]
## Convention updates for CLAUDE.md
- [Specific addition or correction]
## Hook suggestions
- [New hook idea with trigger and behavior]
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.
- 5d ago First seen · 146 lines · 51 tokens per session scan A cfbbfffea0e0
learner is a skill published in the GitHub repository rtazima/claude-proj-blueprint (20 stars, last pushed 3mo ago), licensed MIT. It adds 51 tokens to every session and 1,396 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
learn
Guided, hands-on course teaching architects how to use Codex or Claude Code — six short modules, each built around an exercise on a bundled sandbox project (a fictional Brooklyn art museum expansion). Resumable across sessions via PROGRESS.md. Use when the user runs $learn or /as:learn, says they're new to AI-assisted…
frontmcp-guides
Tutorials, end-to-end walkthroughs, and complete reference projects for FrontMCP. Use when you want a getting-started guide, a full worked example, or to learn best practices by following a step-by-step build rather than a single API reference. Includes a beginner weather-API server (tool plus static resource, Zod…
mandu-explain
Mandu 개념 설명. Island/Filling/Guard/Contract/Slot/SSR/ISR 등 '뭐야'/'어떻게' 시 자동 호출.
svg-press
Teaches agents how to hand-craft self-contained SVG diagrams that render correctly in browsers, tags, and Playwright PDF/PNG. Covers containment math, viewBox sizing, arrow construction, text hierarchy, KDCube brand color tokens, semantic zone fills, badges, the PNG inspection loop, writepng parameters, multi-diagram…
terminal-diagrams
Reference for drawing any diagram as plain monospace terminal text — box-drawing/ASCII, no external renderer. Read when work-preferences.md's trigger fires.
study-buddy
智能督学助手,管理用户的长期学习项目工作流。当用户表达学习项目相关意图时触发:创建/制定学习计划("我想学X"、"帮我制定计划")、汇报学习进度("学完了"、"今天搞定了"、"完成今日任务")、查询计划状态("我学到哪了"、"看下进度")、查看学习报告("日报""周报""月报""项目总结""5月10号到15号的报告")、晨间/晚间打卡复盘、督促、动态调整计划、抱怨学不下去时(情绪支持)。🔴 项目生成流程铁律:项目生成成功后必须一口气走完"项目→知识点→计划表",禁止只汇报"项目已生成 / X 个知识点 / X 个模块"就停下,必须立即输出"DAY / 项目 / 知识点 / 时长 / 难度"表格供用户确认 DAY…