← Back to Skills Marketplace
QMD Learning Loop
by
Shaker Gilbert
· GitHub ↗
· v1.0.1
· MIT-0
118
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install qmd-learning-loop
Description
Capture corrections, errors, feature requests, and recurring best practices in a QMD-native way for markdown-first agent workspaces. Use when a user corrects...
Usage Guidance
This skill is a set of editorial guidelines and templates for capturing and promoting learnings in your markdown workspace. It does not request credentials or install software. The only operational implication is that an agent using this skill will be reading and writing workspace markdown files — so: (1) ensure the agent's file-write permissions are limited to the intended workspace, (2) keep your workspace backed by version control so edits can be reviewed/reverted, (3) review any promoted/created documents for accidental inclusion of secrets or sensitive data, and (4) if you need stricter control, disable autonomous invocation or require manual approval before the agent edits durable docs.
Capability Analysis
Type: OpenClaw Skill
Name: qmd-learning-loop
Version: 1.0.1
The qmd-learning-loop skill bundle establishes a framework for an AI agent to autonomously update its own documentation and behavioral guidelines (e.g., SOUL.md, AGENTS.md, TOOLS.md) based on user interactions and errors. While designed for organizational efficiency and 'durable memory,' the instructions in SKILL.md and references/promotion-targets.md create a high-risk mechanism for persistent prompt injection by allowing external input to be promoted into core operating instructions. There is no evidence of intentional malice, obfuscation, or data exfiltration, but the architectural capability for self-reprogramming based on untrusted user 'corrections' represents a significant security vulnerability.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, and the five reference markdown files all describe and implement the same goal: classify events and route/promote them into workspace docs. There are no unrelated requirements (no env vars, binaries, or external endpoints) that would contradict the stated purpose.
Instruction Scope
Runtime instructions tell the agent to read, classify, and update workspace markdown files (logs, runbooks, decision memos, etc.). That scope is appropriate for a learning-loop skill. The SKILL.md explicitly prefers updating existing docs and keeping chronology separate from durable rules. There is nothing instructing the agent to read unrelated system files, fetch secrets, or transmit data externally.
Install Mechanism
This is an instruction-only skill with no install spec or code files to execute; nothing will be downloaded or written by an installer step. Risk from installation is minimal.
Credentials
No environment variables, credentials, or config paths are requested. The skill operates solely on workspace markdown documents, which is proportionate to its purpose.
Persistence & Privilege
The skill does not request always:true and uses platform defaults (agent-invocable and allowed to invoke autonomously). That is normal for a skill that will edit workspace files. It does not attempt to modify other skills' configs or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install qmd-learning-loop - After installation, invoke the skill by name or use
/qmd-learning-loop - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Broadened skill description to support all markdown-first agent workspaces, not just Atlas/OpenClaw.
- Generalized documentation and routing guidance for greater compatibility with various workspace structures.
- Updated references from specific Atlas/OpenClaw file paths to more general file categories (e.g., "workflow guidance file" instead of "AGENTS.md").
- Simplified examples to avoid assuming the presence of a particular directory structure.
- No code or functional changes; documentation, terminology, and instructions only.
v1.0.0
Initial release of qmd-learning-loop — a QMD-native workflow for capturing and promoting operational learnings:
- Added core workflow and first-/second-pass routing rules for classifying and promoting corrections, errors, feature requests, and best practices.
- Provided specific guidelines for targeting and updating durable docs (e.g., AGENTS.md, TOOLS.md, SOUL.md) without duplicating policy.
- Included criteria for promoting recurring or policy-changing lessons versus logging one-off events in daily memory.
- Added references for routing, promotion, review loop, and starter templates to support the learning capture process.
Metadata
Frequently Asked Questions
What is QMD Learning Loop?
Capture corrections, errors, feature requests, and recurring best practices in a QMD-native way for markdown-first agent workspaces. Use when a user corrects... It is an AI Agent Skill for Claude Code / OpenClaw, with 118 downloads so far.
How do I install QMD Learning Loop?
Run "/install qmd-learning-loop" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is QMD Learning Loop free?
Yes, QMD Learning Loop is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does QMD Learning Loop support?
QMD Learning Loop is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created QMD Learning Loop?
It is built and maintained by Shaker Gilbert (@shakerg); the current version is v1.0.1.
More Skills