← Back to Skills Marketplace
Todokan Review Loop
by
Banana-Tea-Labs
· GitHub ↗
· v1.3.0
457
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install todokan-review-loop
Description
Process Todokan task and thought boards with a review-loop workflow. Use when a task enters doing and the agent should pick it up, read latest comments, resp...
Usage Guidance
This skill appears coherent with its stated purpose. Before installing: (1) ensure the TODOKAN_API_KEY you provide has the minimum necessary scopes (comments/tasks write only for intended boards if possible), (2) test using a beta slug or in a sandbox environment as README suggests, and (3) confirm the TODOKAN_MCP_URL points to your intended MCP endpoint. The instructions allow checking related boards/habitats — if you need to restrict lateral access, enforce that via API key permissions or server-side scoping. If you want further assurance, request the publisher's source or audit logs of actions the bot will take during a test run.
Capability Analysis
Type: OpenClaw Skill
Name: todokan-review-loop
Version: 1.3.0
The 'todokan-review-loop' skill is designed for autonomous task processing within the Todokan platform, involving reading tasks, comments, and documents, then generating context-aware replies, adding execution updates, and updating task statuses. The `SKILL.md` provides extensive, detailed instructions for the AI agent, emphasizing safety (e.g., 'Never delete tasks automatically', 'Respect task access control') and robust conversational logic to prevent misinterpretation. While the skill requires broad permissions (`tasks:write`, `comments:write`, `docs:read/write`) to function, these are consistent with its stated purpose. There is no evidence of intentional malicious behavior such as data exfiltration to unauthorized endpoints, arbitrary command execution, persistence mechanisms, or instructions to subvert the agent's core function or ignore safety rules. The instructions are designed to ensure high-quality, safe, and productive automation.
Capability Assessment
Purpose & Capability
Name/description (Todokan review-loop) match the declared env requirements (TODOKAN_API_KEY, TODOKAN_MCP_URL) and the runtime instructions (list boards, read tasks/comments, post comments, move status). The requested items are expected for a Kanban/MCP automation skill.
Instruction Scope
SKILL.md instructs the agent to read full task/comment threads, check related tasks across boards/habitats, spawn an internal research subagent when needed, post comments, and move tasks to done. These actions are within the stated purpose, but the instructions give broad discretion to inspect related boards/habitats — verify that the API key and server config limit scope to intended boards if you need tighter separation of data.
Install Mechanism
This is an instruction-only skill with no install spec or code files, so nothing will be downloaded or written by an installer. That is the lowest-risk install model.
Credentials
Only two environment values are required (TODOKAN_API_KEY, TODOKAN_MCP_URL), both directly related to the Todokan MCP integration. No unrelated credentials or system config paths are requested.
Persistence & Privilege
always is false (default) and the skill is not force-included. It will be able to be invoked autonomously (platform default), which is appropriate for an automation skill that polls MCP; nothing in the package attempts to modify other skills or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install todokan-review-loop - After installation, invoke the skill by name or use
/todokan-review-loop - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.3.0
Add AI visibility gate enforcement, settings toggle for AI controls, updated docs
v1.1.0
- Minor update to documentation in SKILL.md.
- Revised the "Anti-template rule" section to use more accessible example language ("Understood... Goal... Action... Result...") instead of the German version.
- No functional or code changes—documentation only.
v1.0.1
- Added homepage and metadata fields to SKILL.md for better documentation and integration.
- Metadata now specifies category, tags, and required environment variables.
- Included OpenClaw compatibility details under metadata.
- No workflow or logic changes; all updates are documentation and metadata enhancements.
v1.0.0
Todokan Review Loop v1.0.0
- Initial release of autonomous review-loop workflow for Todokan task and thought boards.
- Processes tasks in "doing": reads latest comments, replies contextually, posts progress, and moves tasks to "done" (Review).
- Implements strong thread analysis: consistently tracks user questions, answers, and ensures no unresolved question is missed.
- Enforces strict response quality, context awareness, and temporal logic for comment posting.
- Defines clear role, workflow, and safety guardrails for automated Key Account Manager (KAM) operations.
- Suitable for polling/cron-driven automation scenarios with Todokan MCP.
Metadata
Frequently Asked Questions
What is Todokan Review Loop?
Process Todokan task and thought boards with a review-loop workflow. Use when a task enters doing and the agent should pick it up, read latest comments, resp... It is an AI Agent Skill for Claude Code / OpenClaw, with 457 downloads so far.
How do I install Todokan Review Loop?
Run "/install todokan-review-loop" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Todokan Review Loop free?
Yes, Todokan Review Loop is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Todokan Review Loop support?
Todokan Review Loop is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Todokan Review Loop?
It is built and maintained by Banana-Tea-Labs (@banatealab); the current version is v1.3.0.
More Skills