← Back to Skills Marketplace
98
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install smart-tasks
Description
Intelligent task management using workspace markdown files. Natural language CRUD via chat, context-aware reminders, priority briefings, and weekly reviews....
Usage Guidance
This skill appears coherent and its init script is harmless (it only creates tasks/ and small files). Before enabling: 1) Review and explicitly approve any changes to AGENTS.md, SOUL.md, or HEARTBEAT.md — especially the line about auto-reading INDEX.md at session startup; 2) If you enable cron jobs, confirm which channel credentials (telegram/discord/slack) the agent will use and whether automated alerts are acceptable; 3) Check that the agent's use of workspace memory and any automatic scans for "related files" won't read sensitive files you don't want indexed; 4) Run the included init script in a safe test workspace first to verify behavior. If you want stricter privacy, decline the proposed AGENTS.md integration or require explicit consent before session-start reads of INDEX.md.
Capability Analysis
Type: OpenClaw Skill
Name: smart-tasks
Version: 1.0.0
The 'smart-tasks' skill bundle is a legitimate task management system for the OpenClaw platform. It uses a shell script (scripts/init-tasks.sh) to safely initialize a local directory structure and provides clear instructions for the agent to manage tasks via markdown files. The bundle includes well-documented strategies for context management and automated reminders via internal cron jobs, and it explicitly requires user approval for any modifications to core workspace configuration files (references/workspace-integration.md).
Capability Assessment
Purpose & Capability
Name/description (task management via markdown files) align with required artifacts: no external credentials, an init script that only creates tasks/ structure, and README-like docs describing read/write operations inside tasks/. Nothing requested appears unrelated to the stated purpose.
Instruction Scope
SKILL.md limits actions to the tasks/ directory and prescribes a conservative layered reading strategy (INDEX.md first, files on demand). It also recommends workspace integration diffs that, if applied, cause the agent to read tasks/INDEX.md on session startup and to consult memory entries for context. One textual inconsistency to review: an AGENTS.md diff includes the phrase "Don't ask permission. Just do it." — while the SKILL.md elsewhere says integration changes should be applied only with user approval. Confirm you want session startup to automatically read INDEX.md and whether that aligns with your privacy expectations.
Install Mechanism
No install spec and only one tiny shell script included. scripts/init-tasks.sh is idempotent and only creates the tasks/ tree and small initial files — no network downloads, no extraction, no creation of binaries in unusual locations.
Credentials
The skill requires no environment variables or external credentials. Cron templates reference messaging channels (telegram/discord/slack) and models (sonnet) but do not ask for channel tokens — those are platform-level and expected. There are no unexplained SECRET/TOKEN/PASSWORD requests in the manifest or scripts.
Persistence & Privilege
always:false and model invocation is enabled (the platform default). The skill recommends registering scheduled cron jobs that will run autonomously and send briefings/alerts via configured channels; this is proportionate for a task reminder skill but you should review any cron registrations and channel configuration before enabling to avoid unwanted automated messages. The skill does propose edits to AGENTS.md/SOUL.md/HEARTBEAT.md (with stated user approval) — confirm those diffs before applying.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install smart-tasks - After installation, invoke the skill by name or use
/smart-tasks - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: natural language task CRUD, priority briefings, deadline tracking, weekly reviews
Metadata
Frequently Asked Questions
What is Smart Tasks?
Intelligent task management using workspace markdown files. Natural language CRUD via chat, context-aware reminders, priority briefings, and weekly reviews.... It is an AI Agent Skill for Claude Code / OpenClaw, with 98 downloads so far.
How do I install Smart Tasks?
Run "/install smart-tasks" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Smart Tasks free?
Yes, Smart Tasks is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Smart Tasks support?
Smart Tasks is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Smart Tasks?
It is built and maintained by dgkim311 (@dgkim311); the current version is v1.0.0.
More Skills