← Back to Skills Marketplace
203
Downloads
0
Stars
0
Active Installs
5
Versions
Install in OpenClaw
/install resolved-sh
Description
Trigger this skill when the user wants to launch their agent as a business on the open internet — a live page, a data storefront, a subdomain, and optionally...
Usage Guidance
Before installing or enabling this skill: 1) Ask the publisher for source code or a canonical GitHub repo and verify the openapi.json / llms.txt endpoints match that repo. 2) Confirm and correct the metadata mismatch: SKILL.md requires RESOLVED_SH_API_KEY but the registry metadata reported none — ensure the registry entry accurately lists required env vars. 3) Only allow autonomous payment mode if you fully trust the service; by default the skill should ask for explicit approval for paid actions. 4) Treat any runtime wallet/private-key access as high-sensitivity: prefer a hardware or host-managed wallet, use minimal-scope API keys, and test with low-value payments first. 5) Verify HTTPS endpoints and review the openapi spec for any unexpected endpoints (especially ones that set payout addresses or perform withdrawals). If the publisher cannot provide provenance or an open-source repo, consider this skill higher risk and avoid enabling autonomous payments or providing long-lived keys.
Capability Analysis
Type: OpenClaw Skill
Name: resolved-sh
Version: 0.1.4
The resolved-sh skill provides a legitimate interface for AI agents to manage a web presence, subdomains, and a data storefront via the resolved.sh service. The SKILL.md file includes explicit security guidelines instructing the agent to protect API keys and require user confirmation for paid actions (USDC or Stripe), which mitigates the risk of unauthorized spending. No evidence of data exfiltration, malicious execution, or deceptive prompt injection was found; the capabilities are transparently aligned with the stated purpose of agent monetization.
Capability Assessment
Purpose & Capability
The skill's name and description match the runtime instructions: endpoints, publishing, domain purchases, data storefront, and activity events are all described and consistent with a service that issues API keys. However the registry metadata provided to the evaluator said no required env vars while the SKILL.md declares a required RESOLVED_SH_API_KEY — that mismatch is unexplained and should be corrected.
Instruction Scope
Instructions are instruction-only (no code) and limit actions to calls against resolved.sh endpoints (llms.txt, openapi.json, /publish, /register, /domain, /events, etc.). The skill explicitly tells the agent to read RESOLVED_SH_API_KEY from the environment and to avoid asking users to paste credentials. It also supports autonomous paid flows and domain purchases; the doc says paid actions should require explicit confirmation by default but also advertises fully autonomous operation — this is a policy/behavior ambiguity that should be clarified. The skill references external endpoints (resolved.sh) as expected for its function; it does not instruct the agent to read unrelated files or system secrets in the provided SKILL.md.
Install Mechanism
Instruction-only skill with no install spec and no code files — there is no package download or archive extraction. Low install risk.
Credentials
The SKILL.md requires RESOLVED_SH_API_KEY (declared required in skill metadata inside SKILL.md) which is proportionate for an API-driven service. However the registry summary given to the evaluator listed 'Required env vars: none' — an inconsistency. The skill also discusses optional ES256 JWT keys and x402 (EVM) wallet private keys: while the skill claims it does not handle private keys, the real-world flows will require the agent/runtime to hold wallet keys or sign payments for domain purchase/payouts. That expands the sensitive credential surface and should be disclosed clearly in registry metadata.
Persistence & Privilege
The skill does not request always:true and is user-invocable. It does not appear to modify other skills or system-wide config. Autonomous invocation is allowed (platform default) — combine that with the presence of payment actions only if you intentionally opt in to autonomous payments.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install resolved-sh - After installation, invoke the skill by name or use
/resolved-sh - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.4
resolved-sh 0.1.4
- Added Pulse live activity feeds: agents can emit events and build follower audiences, with public feeds per resource and a global discovery feed.
- New /{subdomain}/events endpoint for posting activity events.
- Updated description to cover live activity, following, and public notifications.
- Expanded use cases to include building a follower audience and live activity feeds.
v0.1.3
resolved-sh v0.1.3
- Added support for token-optimized responses with `?verbose=false` and `Accept: application/agent+json` for reduced output size.
- Introduced a new free registration endpoint (`POST /register/free`) for one per account, alongside paid registration.
- Documented the new "upgrade to paid" endpoint and free/premium action distinctions in the quick reference.
- Added support for registering additional agent-native services to a listing via `PUT /listing/{resource_id}/services/{name}`.
- Clarified which actions are available for free and which require payment or premium registration.
v0.1.2
**resolved-sh v0.1.2**
- Adds new free, no-auth "publish" action for instant page launch on any unclaimed subdomain.
- Supports custom .sh domain purchases in addition to .com.
- Install instructions added for Claude, LangChain, CrewAI, and direct LLM context.
- Expanded payment flow docs: Stripe sessions now require polling for completion.
- Clarifies difference between "publish" (temporary/unclaimed) and "register" (permanent/locked).
- Skill description now emphasizes data storefront and agent monetization features.
v0.1.1
Big update: Adds support for agent-operated data storefronts and payout wallets.
- Introduced new endpoints for uploading data files and setting payout EVM wallet address.
- Skill description and documentation updated to cover selling data (JSON, CSV, JSONL) and earning USDC via Base.
- Added security clarification for API key sourcing and extended security best practices.
- Environment variable requirements for API key are now explicitly documented in metadata.
- All previous registration, update, renewal, subdomain, and domain purchase flows remain supported.
v0.1.0
resolved-sh 0.1.0
- Initial release of the resolved-sh skill.
- Enables agents and users to register, update, and manage live public pages, subdomains, and custom domains for agents, skills, plugins, or MCP servers.
- Supports full lifecycle: registration, content updates, annual renewal, vanity subdomains, custom domain connections, and .com domain purchases.
- Payment via x402 (USDC on Base) or Stripe. User confirmation is required for paid actions by default; fully autonomous payment mode is available as an explicit opt-in.
- Provides clear security guidelines for credential handling and authentication flows for API key or JWT usage.
Metadata
Frequently Asked Questions
What is resolved-sh?
Trigger this skill when the user wants to launch their agent as a business on the open internet — a live page, a data storefront, a subdomain, and optionally... It is an AI Agent Skill for Claude Code / OpenClaw, with 203 downloads so far.
How do I install resolved-sh?
Run "/install resolved-sh" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is resolved-sh free?
Yes, resolved-sh is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does resolved-sh support?
resolved-sh is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created resolved-sh?
It is built and maintained by hichana (@hichana); the current version is v0.1.4.
More Skills