- Home
- Skills
- Productivity
- Agent Task Hub
Agent Task Hub
File-based Markdown task queue for a human and coding agent: priorities, dependencies, journal, explicit start only.
Free
Works with the AI Agents you already use
Agent Task Hub
Example session with this skill installed
Show me the current task hub queue and take the next P1 task that is ready to work on.
- Read your context and instructions
- Compiled the agent task hub
I have audited the task queue in the tasks/ directory. There are currently 3 tasks ready, 1 in progress, and 1 recurring task suggested.
Task Queue Status
| ID | Priority | Status | Task Name | Dependencies |
|---|---|---|---|---|
| 0042 | P1 | ready | Implement OAuth2 provider | 0039 |
| 0045 | P2 | ready | Refactor database connection pool | None |
| 0048 | P3 | ready | Update README documentation | None |
| 0041 | P1 | in_progress | Fix memory leak in worker | None |
| R0001 | P2 | ready | Weekly dependency audit | None |
I am continuing the current P1 task: 0041: Fix memory leak in worker.
Next steps
- Review the latest entry in the Journal and continuation section of
tasks/0041.md. - Execute the reproduction script to verify the leak is still present.
- Update the status to
doneonce the heap profile confirms the fix.
Connects securely to your tools. The creator never sees your data.
About this skill
Manage a persistent, file-based task queue designed specifically for collaboration between humans and AI coding agents. This skill eliminates the confusion of lost context and unorganized backlogs by using Markdown files as the single source of truth for project state.
What it does
- Queue management reads and writes to a structured
tasks/*.mddirectory to maintain a shared state. - Priority sorting selects the next logical action based on P1 to P3 rankings and satisfied dependencies.
- State tracking manages task lifecycles from
readytoin_progress,blocked, ordone. - Recurring workflows handles procedures like maintenance or backups through on-demand recurring task templates.
- Journaling maintains a dated log of progress and results within each task file for easy handoffs.
How it works
- Queue Audit: Use the "show queue" command to scan all task files and validate the current backlog state.
- Task Ingestion: Add new tasks to the directory using the standard or recurring Markdown templates.
- Execution: Command the agent to "take the next task" to automatically identify and start the highest priority ready item.
- Lifecycle Update: The agent updates the journal, results, and status fields before rebuilding the local dashboard.
Frameworks & tools
This skill requires a local Markdown-based task directory. It integrates with optional Python 3.10+ scripts for validation and dashboard generation via make commands and uses local Git for state commits.
Why this beats prompting it yourself
Generic prompts often lose track of what has been finished or what is currently blocked across chat sessions. This skill enforces a strict file-based protocol that ensures the agent never starts unauthorized work and always respects complex dependency chains.
Use cases
- Manage technical debt by queuing P3 refactoring tasks during active development.
- Coordinate handoffs between human coding sessions and autonomous agent runs.
- Automate weekly maintenance scripts through the recurring task framework.
- Resolve blockers by explicitly marking tasks as blocked to stop agent execution.
Known limitations
Requires a specific directory structure and file schema to function. It does not support automatic cloud syncing or pushing to remote repositories without manual confirmation.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Productivity.
Requires specific tasks/ directory structure and templates. Agent will not auto-start tasks without explicit commands. Python 3.10+ needed for optional dashboard and validation features.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 15 days ago
- Free with an Agensi account
Permissions
File Scopes