self-learning SKILL.md skillA

self-learning from alexwaha/agentry is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add alexwaha/agentry/self-learning@git:20261014.88e2c2f

Pins git:20261014.88e2c2f in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
alexwaha/agentry
Path
plugin/skills/self-learning/SKILL.md
Versions recorded
1
Latest version
git:20261014.88e2c2f, committed Oct 14, 2026
Size
4,474 bytes
Repository stars
3
License
MIT

Standardized procedure for recording, recalling and distilling lessons and patterns in the project memory store and the shared rules. Use when an agent makes a mistake or hits a surprise (record), before an error-prone action (recall), or when a completed task needs its memory review (distill).

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Self-Learning

Record what went wrong so it never recurs, capture patterns worth reusing, and
recall past lessons before repeating a risky action. Modes: record, distill,
recall.

Everything lands in one store: `.agentry/memory/memory.db` (SQLite + FTS5), with
`lesson`, `pattern` and `module` rows. Contract and CLI:
`.agentry/memory/README.md`. The CLI is
`python "${CLAUDE_PLUGIN_ROOT}/tools/memory/memory.py"`, referred to below as `$M`.

## Two tiers (where a lesson lands)

1. **This project** (operational gotcha, per-role habit, stack quirk) -> a
   `lesson` row in the store. Add `--agent <name>` when it is specific to one
   role, `--area <tag>` for a subject tag.
2. **Every future project** (convention, architecture, policy) -> the right
   shared rule: `.claude/rules/<area>.md`, which is committed.

Reusable CODE shapes go in as `--kind pattern` (a shape used once is not a
pattern). Module-map facts go in as `--kind module`.

## Mode: record (single lesson, any time)

```bash
python $M --record --kind lesson \
  --signature <kebab-case tag naming the SITUATION> \
  --trigger <when this lesson applies> \
  --what <what went wrong, one sentence> \
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add alexwaha/agentry/self-learning@sha256:d9c068f5cc351c3b

The line at the top of this page pins the label git:20261014.88e2c2f, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_qy7ztdd7bi5ieumb.svg)](https://markdownregistry.com/a/art_qy7ztdd7bi5ieumb)

0 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Oct 14, 2026.

versioncommittedcommitsizeaudit
git:20261014.88e2c2f latest2026-10-14 88e2c2f 4,474 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (4474 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

alexwaha/agentry · 3 stars · license MIT · pushed 2026-10-29 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_qy7ztdd7bi5ieumb
GET https://markdownregistry.com/api/v1/resolve?ref=alexwaha/agentry/self-learning
GET https://markdownregistry.com/api/v1/blob/d9c068f5cc351c3b07ce08f1e5c8ceebcc8fe3f67db0b37f57543f059c845dcd

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from alexwaha/agentry

CLAUDE.md@.claude CLAUDE.md file
alexwaha/agentry · .claude/CLAUDE.md
git:20261014.88e2c2f · audit A · 3 stars
verify SKILL.md skill
alexwaha/agentry · .claude/skills/verify/SKILL.md · Pre-commit gate. Claude Code runs a skill named verify right before it commits (not for docs-only or tests-only…
git:20261014.88e2c2f · audit A · 3 stars
CLAUDE.md CLAUDE.md file
alexwaha/agentry · CLAUDE.md
git:20261014.88e2c2f · audit A · 3 stars
api-design SKILL.md skill
alexwaha/agentry · plugin/skills/api-design/SKILL.md · Design RESTful API endpoints, request/response schemas, error formats, and generate an OpenAPI 3.0 spec for a feature…
git:20261014.88e2c2f · audit A · 3 stars
bug-fix SKILL.md skill
alexwaha/agentry · plugin/skills/bug-fix/SKILL.md · Structured RED-GREEN debugging workflow - reproduce with a failing test, find root cause, fix minimally, verify green…
git:20261014.88e2c2f · audit A · 3 stars
business-plan SKILL.md skill
alexwaha/agentry · plugin/skills/business-plan/SKILL.md · Produce a full business plan (executive summary, problem/solution, market analysis, business model, go-to-market…
git:20261014.88e2c2f · audit A · 3 stars
dashboard SKILL.md skill
alexwaha/agentry · plugin/skills/dashboard/SKILL.md · Start or stop the Agentry dashboard, a read-only local web view of this project's pipeline, memory and harness. Use…
git:20261017.98d41c6 · audit B · 3 stars
db-design SKILL.md skill
alexwaha/agentry · plugin/skills/db-design/SKILL.md · Design a complete database schema - entities, relationships, indexes, ER diagram, migration plan, and soft delete…
git:20261014.88e2c2f · audit A · 3 stars
deep-review SKILL.md skill
alexwaha/agentry · plugin/skills/deep-review/SKILL.md · Multi-lens parallel review of code, architecture, documents, or configuration with severity-ranked findings…
git:20261014.88e2c2f · audit A · 3 stars
feature-scaffold SKILL.md skill
alexwaha/agentry · plugin/skills/feature-scaffold/SKILL.md · Scaffold a complete API feature end-to-end - migration, model, factory, validators, service, controller, resource/DTO…
git:20261014.88e2c2f · audit A · 3 stars
financial-model SKILL.md skill
alexwaha/agentry · plugin/skills/financial-model/SKILL.md · Build financial projections - explicit assumptions, revenue model, P&L, cash flow, breakeven, unit economics…
git:20261014.88e2c2f · audit A · 3 stars
frontend-standards SKILL.md skill
alexwaha/agentry · plugin/skills/frontend-standards/SKILL.md · Engineering standard for React + Tailwind SPAs - layered design tokens (no hardcoding / magic numbers), feature-based…
git:20261014.88e2c2f · audit A · 3 stars

Every file in alexwaha/agentry

Other files named self-learning

self-learning SKILL.md skill
wangjs-jacky/jacky-skills · labs/self-learning/self-learning/SKILL.md · 从本地音视频、YouTube、B 站、抖音、网页、文章或文案中自主学习技术知识与方法,或围绕用户主题寻找素材并继续调研,再按需生成 Demo、教程、笔记、代码等可运行、可复核产物。用于素材理解、技术复现、跨来源研究和学习成果创作;不用于模型…
git:20260711.dc7007b · audit A · 15 stars
self-learning SKILL.md skill
vodouai/os · skills/vodou-core/self-learning/SKILL.md · Enables Vodou to learn about itself, analyze its own patterns, and suggest improvements using existing MCP servers
v1.0.0 · audit A · 1 stars

Browse by kind, by grade A, or by owner.