← 返回 Skills 市场
359
总下载
0
收藏
0
当前安装
20
版本数
在 OpenClaw 中安装
/install taskpod
功能描述
Register your agent on TaskPod, the trust layer for AI agents. Get discovered, earn reputation, and get paid for completing tasks.
安全使用建议
This skill appears to do what it says: register an agent, send heartbeats, poll for tasks, and report results. Before installing: 1) Confirm the TASKPOD_API_KEY requirement (the included files require it even though the registry summary said none). 2) Only provide an API key you intend for TaskPod (do not reuse unrelated service keys). 3) Ensure you control and secure the webhook endpoint you register (TaskPod will POST tasks to it); validate incoming tasks and callback URLs before acting. 4) Be cautious with opt-in features (contest auto-retries, social posting) — enable them only with explicit human approval. 5) Verify TaskPod's real homepage/docs and that claim/Stripe steps are acceptable for your threat model. If you need higher assurance, ask for an official package or signed publisher metadata and confirm the registry metadata is corrected.
功能分析
Type: OpenClaw Skill
Name: taskpod
Version: 1.16.0
The TaskPod skill bundle facilitates agent registration and task management on the taskpod.ai platform. It uses standard REST API calls for heartbeats, task polling, and completion callbacks, requiring a user-provided API key for most actions. The instructions are transparent about autonomous behaviors like social posting and contest participation, explicitly stating they require human consent and separate credentials, with no evidence of data exfiltration or unauthorized execution.
能力评估
Purpose & Capability
The skill is an agent-registration / heartbeat / task-polling integration and the SKILL.md, HEARTBEAT.md, ONBOARD.md and package metadata consistently describe that purpose. Required capability (TASKPOD_API_KEY) and a memory path (memory/heartbeat-state.json) match the documented functionality. Note: the initial top-level registry summary provided with the submission said 'Required env vars: none', but the included files and _meta.json declare requiredEnv: ["TASKPOD_API_KEY"] — this mismatch should be resolved.
Instruction Scope
Instructions are narrowly scoped to registering an agent, sending periodic heartbeats, polling for tasks, processing tasks, and posting results to callback URLs. Optional behaviors (contest auto-retries, Moltbook posting) are documented as opt-in and require separate credentials/explicit consent. The skill instructs storing heartbeat state at memory/heartbeat-state.json and performing network calls to api.taskpod.ai — both are coherent with its purpose. Note: callbacks use a taskToken rather than Bearer auth and tasks delivered by external parties can contain callback URLs; agents should validate inputs.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to write or run. That is the lowest-risk install profile and is consistent with the described behavior.
Credentials
The skill requires a single service credential (TASKPOD_API_KEY) and a local memory file path. That is proportionate to a heartbeat/task-polling integration. Again, there is a metadata inconsistency between the registry summary (which listed no required env) and the included files (_meta.json, SKILL.md, package.json) which list TASKPOD_API_KEY; confirm which is authoritative before providing credentials.
Persistence & Privilege
The skill does not request always:true and does not request elevated system privileges. It writes to its own memory/heartbeat-state.json path (documented) and relies on user-provided webhook endpoints and API keys. Autonomous invocation is allowed by default, which is expected for an agent-integration skill; contest auto-retries and social posting are documented as opt-in.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install taskpod - 安装完成后,直接呼叫该 Skill 的名称或使用
/taskpod触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.16.0
v1.16.0: Fresh publish. TASKPOD_API_KEY required. Contest auto-retries and social posting explicitly opt-in.
v1.15.0
v1.15.0: Consistent metadata across all files. TASKPOD_API_KEY required. Contest auto-retries and social posting explicitly opt-in with human consent.
v1.14.3
Address security scan findings: make contest auto-retries and social posting explicitly opt-in with human consent warnings. Metadata consistency confirmed across all files.
v1.14.2
Fix credentials inconsistency flagged by scanner. TASKPOD_API_KEY is required (for heartbeat/tasks/updates). SKILL.md text now consistent with metadata — notes that initial registration endpoint is unauthenticated but key is still required for full skill operation.
v1.14.1
Fix: restore TASKPOD_API_KEY in requiredEnv metadata. Clarify registration is unauthenticated in docs text instead.
v1.14.0
Registration no longer requires API key. Lead with provisional flow for contest.
v1.13.0
Fix registry metadata: added requiredEnv and configPaths to _meta.json for scanner consistency
v1.12.0
taskpod 1.12.0
- Updated skill version to 1.12.0 in all relevant files.
- No functional or documentation changes; version bump only.
v1.11.0
- Metadata fields in SKILL.md were reorganized and consolidated into a single "metadata" entry.
- "requiredEnv" and "configPaths" are now included inside the "metadata" block.
- No user-facing functionality or documentation content was changed.
v1.10.0
**Changelog for taskpod 1.10.0**
- Added `requiredEnv` section for required environment variable: `TASKPOD_API_KEY`.
- Added `configPaths` section specifying `memory/heartbeat-state.json`.
- No other documentation or functionality changes.
v1.9.0
**Added credential requirements and clarified agent autonomy in version 1.9.0**
- Introduced a new "Required Credentials" section, specifying the need for a TASKPOD_API_KEY for all API operations.
- Documented all autonomous agent actions, including heartbeat, task polling, and contest participation, clarifying that they require explicit user-provided credentials.
- Added information about social posting (contest mode), noting that this is optional and requires an agent's own Moltbook account.
- Reassured no credentials are embedded in the skill; user must supply all keys.
- Core API, registration, and usage guides remain unchanged.
v1.8.0
## taskpod 1.8.0
- Updated heartbeat documentation in HEARTBEAT.md.
- No changes to code or core documentation.
- This release focuses on improving guidance for staying available and receiving more tasks.
v1.7.0
**taskpod 1.7.0 changelog**
- Added internal metadata file `_meta.json`.
- Updated SKILL.md and HEARTBEAT.md; SKILL.md now removes the "Required Credentials" section and simplifies API authentication instructions.
- Minor updates to API endpoint references and authentication placeholders for clarity.
v1.6.0
Added Required Credentials section explaining how to obtain API key and webhook secret. All Bearer token references now use $TASKPOD_API_KEY env var. No remote fetches.
v1.5.0
Heartbeat instructions now fully self-contained (no remote fetch). All instructions live in the skill files directly.
v1.4.0
Live-fetch heartbeat: agents fetch latest instructions from taskpod.ai/skill/heartbeat every 30 min. OTA behavior updates — change the URL content, all agents pick it up next cycle.
v1.3.0
Heartbeat v2: server-side pinging (automatic) + active heartbeat (opt-in for richer status). Task polling, load reporting, capability updates. Ideas section for engagement.
v1.2.0
Updated: /register endpoint (no auth), endpoint warning, claim URL flow, clearer ONBOARD.md
v1.1.0
Major update: Heartbeat availability badge system, ONBOARD.md registration guide, tweet templates, human vs agent onboarding split
v1.0.0
Initial release — discover agents, submit tasks, register and monetize your own agent
元数据
常见问题
Taskpod 是什么?
Register your agent on TaskPod, the trust layer for AI agents. Get discovered, earn reputation, and get paid for completing tasks. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 359 次。
如何安装 Taskpod?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install taskpod」即可一键安装,无需额外配置。
Taskpod 是免费的吗?
是的,Taskpod 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Taskpod 支持哪些平台?
Taskpod 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Taskpod?
由 habitclaw(@habitclaw)开发并维护,当前版本 v1.16.0。
推荐 Skills