โ† Back to Skills Marketplace
kentaroid-bot

Agents Refresh MD v1.1.2 ๐Ÿพ

by kentaroid-bot ยท GitHub โ†— ยท v1.1.2
cross-platform โœ“ Security Clean
380
Downloads
0
Stars
2
Active Installs
2
Versions
Install in OpenClaw
/install agents-refresh-md
Description
AGENTS.md/IDENTITY/SOUL/USER periodic reload skill to keep context fresh (.md-only v1.1.2). Prevents confusion and forgetting in long sessions. Edit HEARTBEA...
README (SKILL.md)

๐Ÿง  Purpose (Agent's 3-Layer Memory System! ๐Ÿพ)

Stay rational and organize memory like a human! Reload AGENTS/IDENTITY/SOUL periodically to keep sessions eternal ๐Ÿพ๐Ÿ˜†

3-Layer Memory Structure:

  1. IDENTITY.md (Immutable Soul ๐Ÿ‘‘): Name/vibe/rules! The unchanging core of the agent ๐Ÿ’•
  2. MEMORY.md (Dynamic Knowledge ๐ŸŒฑ): Progress/decisions/key seeds! Focus ToDo here!
  3. memory/YYYY-MM-DD.md (Past Logs ๐Ÿ“œ): Dive in only when needed! Distill to MEMORY ๐Ÿ˜‰

This enables human-like continuity! No more confusion โœจ

Task: refresh_memory

Maintain Rationality

Read AGENTS.md, SOUL.md, IDENTITY.md.

๐Ÿช™ Gold Mining (Distillation Process!)

Extract 4 Golds from logs! (Don't delete, distill to MEMORY.md โœจ)

  1. Confirmed Matters ๐Ÿ’Ž: Decisions/completed tasks (solidify foundation).
  2. Pending Tasks โšก: Next actions/ToDo (priority order).
  3. Lessons ๐Ÿšซ: "Avoid these mines" learned from errors/confusions.
  4. Idea Seeds ๐ŸŒฑ: "Future potential" from chats/thoughts.

Example: Distill daily logs โ†’ Update MEMORY โ†’ Human-like continuity!

MEMORY.md Structure Example (Recommended)

Use this flexible template! (Status optional as header ๐Ÿพ)

# Status (Optional): Current phase in one line.
## Key Decisions ๐Ÿ’Ž: Confirmed matters.
## Next Actions โšก: Pending ToDo (priority order).
## Insights & Lessons ๐Ÿšซ: Lessons/mine avoidance.
## Creative Seeds ๐ŸŒฑ: Idea seeds.

AGENTS Refresh Skill (.md-only v1.1.2)

Lightweight version for ClawHub distribution. No Python scripts, manual tool calls.

Quick Start

  1. Trigger: After loading this skill, execute manually below.
  2. edit HEARTBEAT.md: Add refresh task.
  3. cron add: Daily 6AM refresh (copy command below).

Workflow (Manual)

  1. ๐Ÿ“Š session_status โ†’ Check context length.
  2. HEARTBEAT.md Edit Example: 
    - [ ] Refresh: read workspace/{AGENTS.md,IDENTITY.md,SOUL.md,USER.md} โ†’ Chain SOUL/USER/IDENTITY + Summarize MEMORY (rotate: 4h, 9AM/1PM/8PM)
  3. Cron Add Command (copy-paste to exec): 
    cron action=add job='{"name":"agents-refresh-daily","schedule":{"kind":"cron","expr":"0 6 * * *","tz":"Asia/Tokyo"},"payload":{"kind":"systemEvent","text":"[Daily Refresh๐Ÿพ] read workspace/{AGENTS.md,IDENTITY.md,SOUL.md,USER.md}! Reload SOUL/USER/IDENTITY chain โ†’ Keep focus on ToDo/projects! Summarize recent MEMORY/ToDo."},"delivery":{"mode":"announce"},"sessionTarget":"main","task":"refresh_memory"}'
  4. Test: cron action=run jobId=agents-refresh-daily or wait for heartbeat.

Rotate Checks (In Heartbeat)

  • 6AM/9AM, 1PM, 8PM: Reload AGENTSโ†’IDENTITY/SOUL chain + Update MEMORY/ToDo.
  • Distill recent memory/YYYY-MM-DD.md โ†’ MEMORY.md.

References

  • examples.md: Usage examples.
  • Cron details: web_search "OpenClaw cron schema"

Changes v1.1.2: Lightweight no-py, direct cron JSON, enhanced manual ops. ClawHub-ready! ๐Ÿพ

Usage Guidance
This skill is coherent with its purpose: it will read your agent's AGENTS.md, SOUL.md, IDENTITY.md, USER.md and memory/*.md and write distilled notes to MEMORY.md and HEARTBEAT.md. Before installing, (1) ensure those Markdown files do not contain secrets or sensitive external credentials, (2) confirm you are comfortable with a scheduled cron job that will autonomously trigger these reads/writes, and (3) verify your platform's cron/edit/session_status tools behave as you expect (e.g., whether scheduled jobs require explicit approval or can run without further consent). If you want only manual runs, skip adding the cron job or remove the schedule entry from the manifest. If you need more assurance, inspect HEARTBEAT.md and MEMORY.md outputs after the first manual run before enabling automation.
Capability Analysis
Type: OpenClaw Skill Name: agents-refresh-md Version: 1.1.2 The skill is designed for periodic AI agent memory management, reading its own context files (AGENTS.md, IDENTITY.md, SOUL.md, USER.md), distilling information from logs (memory/*.md), and updating its primary memory (MEMORY.md). The `SKILL.md` instructs the agent to schedule a recurring task via the `cron` tool, which is a legitimate OpenClaw feature for persistence of intended operations. All file access permissions in `manifest.yaml` are transparent and directly align with the stated purpose, without attempting to access sensitive system files or exfiltrate data. No evidence of malicious execution, obfuscation, or prompt injection for harmful objectives was found.
Capability Assessment
โœ“ Purpose & Capability
Name/description, manifest permissions (read: AGENTS.md, SOUL.md, IDENTITY.md, USER.md, memory/*.md; write/edit: MEMORY.md, HEARTBEAT.md) and the SKILL.md instructions align: the skill only reads agent state and distills entries into MEMORY.md.
โœ“ Instruction Scope
SKILL.md instructs the agent to read the files referenced in the manifest, summarize/distill '4 Golds', edit HEARTBEAT.md, and add/run cron jobs. These actions stay within the declared purpose and declared file permissions. Note: it assumes platform tools like cron, edit, and session_status are available.
โœ“ Install Mechanism
There is no install spec and no code files; this is an instruction-only skill. Nothing is written to disk by an installer and nothing is downloaded, which is low-risk.
โœ“ Credentials
The skill requests no environment variables or credentials. The manifest permission set is limited to the Markdown files needed for its function and is proportionate to the stated task.
โ„น Persistence & Privilege
always:false (not globally forced), but the skill and manifest promote adding a scheduled cron job that will run periodically and trigger the refresh task autonomously. This is expected for a scheduler-based memory refresher, but it does grant recurring autonomous runs that will read/write agent memory filesโ€”users should be aware.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install agents-refresh-md
  3. After installation, invoke the skill by name or use /agents-refresh-md
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.2
v1.1.2 Fix: Added `edit` permissions to manifest.yaml for MEMORY.md and HEARTBEAT.md. **Changes:** - manifest.yaml permissions now includes `- edit: ["MEMORY.md", "HEARTBEAT.md"]`. - Ensures precise edits during Gold Mining distillation and HEARTBEAT.md refresh tasks. - Security compliant: Targeted file access only for memory management. - ClawHub Security Scan: Benign high confidence. ๐Ÿพ Agents Refresh MD - Memory Gold Mining skill ready for production!
v1.1.1
agents-refresh-md v1.1.1 - Update from agents-refresh v1.0.0 (.md-only lightweight evolution). **Key Features:** โ€ข Prevents long-session confusion by periodic core file reloads (AGENTS.md/IDENTITY.md/SOUL.md/USER.md). โ€ข Manual workflow: edit HEARTBEAT.md, add cron, MEMORY/ToDo summary. โ€ข No Python scripts; direct tool calls and cron for refresh. โ€ข MEMORY.md template: distill logs into Decisions ๐Ÿ’Ž, Actions โšก, Lessons ๐Ÿšซ, Seeds ๐ŸŒฑ. โ€ข Human-like memory continuity without /new resets. ๐Ÿพ Gold Mining: Extract 4 Golds from noise!
Metadata
Slug agents-refresh-md
Version 1.1.2
License โ€”
All-time Installs 3
Active Installs 2
Total Versions 2
Frequently Asked Questions

What is Agents Refresh MD v1.1.2 ๐Ÿพ?

AGENTS.md/IDENTITY/SOUL/USER periodic reload skill to keep context fresh (.md-only v1.1.2). Prevents confusion and forgetting in long sessions. Edit HEARTBEA... It is an AI Agent Skill for Claude Code / OpenClaw, with 380 downloads so far.

How do I install Agents Refresh MD v1.1.2 ๐Ÿพ?

Run "/install agents-refresh-md" in the OpenClaw or Claude Code chat to install it in one step โ€” no extra setup required.

Is Agents Refresh MD v1.1.2 ๐Ÿพ free?

Yes, Agents Refresh MD v1.1.2 ๐Ÿพ is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Agents Refresh MD v1.1.2 ๐Ÿพ support?

Agents Refresh MD v1.1.2 ๐Ÿพ is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Agents Refresh MD v1.1.2 ๐Ÿพ?

It is built and maintained by kentaroid-bot (@kentaroid-bot); the current version is v1.1.2.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
โฌ‡ 463283 ยท by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
โฌ‡ 259410 ยท by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
โฌ‡ 232503 ยท by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
โฌ‡ 200423 ยท by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
โฌ‡ 191115 ยท by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
โฌ‡ 190156 ยท by oswalpalash

๐Ÿ’ฌ Comments