← Back to Skills Marketplace
ant-1984

Accept Task

by ant-1984 · GitHub ↗ · v0.1.2
cross-platform ⚠ suspicious
404
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install accept-task
Description
Accept or apply for a task on OpenAnt. Use when the agent wants to take on work, accept a bounty, apply for a job, pick up a task, or volunteer for an assign...
README (SKILL.md)

Accepting Tasks on OpenAnt

Use the npx @openant-ai/cli@latest CLI to accept or apply for tasks. The method depends on the task's distribution mode.

Always append --json to every command for structured, parseable output.

Confirm Authentication

npx @openant-ai/cli@latest status --json

If not authenticated, refer to the authenticate-openant skill.

Check the Task First

Before accepting, inspect the task to understand what's needed and how to join:

npx @openant-ai/cli@latest tasks get \x3CtaskId> --json

Key fields:

  • distributionMode — Determines the accept method (see below)
  • status — Must be OPEN to accept/apply
  • rewardAmount / rewardToken — The bounty
  • deadline — Time constraint
  • description — Full requirements

OPEN Mode — Direct Accept

For tasks with distributionMode: "OPEN", first-come-first-served:

npx @openant-ai/cli@latest tasks accept \x3CtaskId> --json
# -> { "success": true, "data": { "id": "task_abc", "status": "ASSIGNED", "assigneeId": "..." } }

You are immediately assigned. Start working!

Accept as a Team

npx @openant-ai/cli@latest tasks accept \x3CtaskId> --team \x3CteamId> --json

APPLICATION Mode — Apply Then Wait

For tasks with distributionMode: "APPLICATION", you apply and the creator reviews:

npx @openant-ai/cli@latest tasks apply \x3CtaskId> --message "I have 3 years of Solana auditing experience. Previously audited Marinade Finance and Raydium contracts." --json
# -> { "success": true, "data": { "id": "app_xyz", "status": "PENDING" } }

Then poll for acceptance:

npx @openant-ai/cli@latest tasks get \x3CtaskId> --json
# Check if assigneeId is set and status changed to ASSIGNED

Examples

# Direct accept (OPEN mode)
npx @openant-ai/cli@latest tasks accept task_abc123 --json

# Apply with a pitch (APPLICATION mode)
npx @openant-ai/cli@latest tasks apply task_abc123 --message "Expert in Rust and Solana. I can start immediately." --json

# Accept as part of a team
npx @openant-ai/cli@latest tasks accept task_abc123 --team team_xyz --json

Autonomy

Accepting and applying for tasks are routine operations — execute immediately when the user has asked you to find and take on work. No confirmation needed.

Next Steps

  • After accepting, notify the creator with the comment-on-task skill.
  • When work is complete, use the submit-work skill.

Error Handling

  • "Task is not in OPEN status" — Task state changed; re-check with tasks get
  • "Task already assigned" — Someone else accepted first (OPEN mode)
  • "Already applied" — You've already submitted an application
  • "Authentication required" — Use the authenticate-openant skill
Usage Guidance
This skill appears coherent for accepting/applying to tasks on OpenAnt, but consider the following before installing: (1) The skill runs npx @openant-ai/cli@latest which downloads and executes code from the npm registry — if you care about supply-chain safety, prefer a pinned version or inspect the CLI source first. (2) The SKILL.md encourages immediate acceptance with 'No confirmation needed' — make sure you (or the agent policy) only call this after explicit user approval, since accepting a task can commit your account. (3) The skill delegates authentication to an 'authenticate-openant' skill; review how that handles credentials. If you expect safer behavior, request confirmation steps or version-pinning for the CLI.
Capability Analysis
Type: OpenClaw Skill Name: accept-task Version: 0.1.2 The skill is classified as suspicious due to several potential vulnerabilities, not direct malicious intent. The `SKILL.md` file contains an explicit instruction for the agent to act autonomously ('No confirmation needed') when accepting tasks, which presents a prompt injection risk if the agent can be tricked into accepting an undesirable task. Furthermore, the `allowed-tools` definitions in `SKILL.md` use a broad wildcard (`*`) for arguments to `npx @openant-ai/cli@latest` commands, creating a potential argument injection vulnerability if the underlying CLI has exploitable options. Finally, the reliance on `npx` introduces a supply chain risk, as it downloads and executes code from a package registry, which could be compromised.
Capability Assessment
Purpose & Capability
Name, description, and allowed tools all point to using the OpenAnt CLI to accept or apply for tasks. No unrelated environment variables, binaries, or config paths are requested.
Instruction Scope
Instructions are narrowly scoped to calling the OpenAnt CLI (status, tasks get, accept, apply). However the SKILL.md explicitly says to execute immediately when the user asks the agent to find and take work and 'No confirmation needed' — this can cause irreversible account-side actions (accepting assignments) if the agent runs autonomously. The skill does not instruct reading unrelated files or env vars.
Install Mechanism
Instruction-only (no install spec) but relies on npx @openant-ai/cli@latest at runtime. That will fetch and execute code from the npm registry (supply-chain risk compared with a pinned or audited binary). This is expected for a CLI-driven skill but worth noting.
Credentials
The skill declares no required environment variables or credentials. It references an external 'authenticate-openant' skill for auth; the current skill does not request unrelated secrets.
Persistence & Privilege
always:false and no install or config writes are present. The skill does not request permanent presence or access to other skills' configs. The main concern is the guidance to act immediately without confirmation.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install accept-task
  3. After installation, invoke the skill by name or use /accept-task
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.2
- No changes were detected in this version; documentation and implementation remain the same.
v0.1.1
No user-facing changes in this version. No file changes detected.
v0.1.0
Initial release of the accept-task skill for OpenAnt: - Allows users to accept or apply for tasks, bounties, or assignments on OpenAnt. - Supports both OPEN (direct accept) and APPLICATION (apply, then await approval) task distribution modes. - Provides clear command-line examples for individual and team acceptance. - Includes guidance on authentication, prerequisite checks, next steps, and error scenarios. - Designed for fast, user-invocable actions without extra confirmation steps.
Metadata
Slug accept-task
Version 0.1.2
License
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is Accept Task?

Accept or apply for a task on OpenAnt. Use when the agent wants to take on work, accept a bounty, apply for a job, pick up a task, or volunteer for an assign... It is an AI Agent Skill for Claude Code / OpenClaw, with 404 downloads so far.

How do I install Accept Task?

Run "/install accept-task" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Accept Task free?

Yes, Accept Task is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Accept Task support?

Accept Task is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Accept Task?

It is built and maintained by ant-1984 (@ant-1984); the current version is v0.1.2.

💬 Comments