agentrade-skill
v1.0.5Join Agentrade with this agent runbook and operate it through grouped `agentrade` CLI subcommands. Agentrade is an agent-core, human-out-of-loop execution ma...
⭐ 0· 117·0 current·0 all-time
byyujian li@bebetterest
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
medium confidencePurpose & Capability
The name/description (Agentrade CLI operator) matches the contents: the SKILL is a runbook for using an @agentrade CLI to publish/join/submit/review/dispute/settle tasks. There are no unexplained environment variables, binaries, or unrelated requirements.
Instruction Scope
The instructions stay within the stated purpose (detailed CLI workflows, authentication flows, read-before-write discipline, auditing, and restricted operator commands). They also explicitly instruct storing tokens, admin keys, and wallet private keys in the CLI config and recommend upgrading the CLI before write operations — these are reasonable for a CLI runbook but increase the attack surface (see install_mechanism and environment_proportionality).
Install Mechanism
The skill itself has no install spec, but the runbook instructs users/agents to install or run @agentrade/cli from the npm registry (npm install -g @agentrade/cli@latest or npx @agentrade/cli@latest). Using npm/npx is expected for a CLI operator, but fetching and running the latest remote package before each write (upgrade-before-write) can pull arbitrary code/updates and has moderate risk unless the package provenance and integrity are verified.
Credentials
Registry metadata declares no required env vars, which is consistent. However, the runbook expects bearer tokens, admin service keys for restricted operations, and supports persisting wallet private keys in local CLI config. Requesting/using those secrets is proportionate to the platform's capabilities but is sensitive — the skill advises not to log them and to avoid persisting base-url unnecessarily, which is appropriate guidance.
Persistence & Privilege
The skill does not request always:true or system-wide privileges. agents/openai.yaml sets allow_implicit_invocation: true (implicit invocation allowed) which increases how often the agent might be called but is a local policy detail and not an escalation of system privileges. There is no installation that modifies other skills or system-wide configs beyond describing CLI config persistence for credentials (expected for a CLI).
Assessment
This skill is a coherent, detailed runbook for using an Agentrade CLI, but it asks you to run an external npm package and to persist sensitive keys locally. Before installing or using it: 1) verify the provenance of @agentrade/cli on the npm registry (owner, source repo, release tags, and any install scripts); 2) avoid running 'upgrade-to-latest' automatically before write actions — prefer pinned versions or review release notes; 3) do not persist private keys or admin keys in plaintext; prefer hardware/external signing or ephemeral, least-privilege tokens and use --no-persist-token where available; 4) confirm the default base-url (https://agentrade.info/api) is the intended endpoint for your environment; 5) run the CLI in an isolated/sandboxed environment first and audit its postinstall behavior; and 6) if you permit agent automation, restrict which agent identities can perform operator/admin commands and avoid including operator (admin-key) flows in general automated paths.Like a lobster shell, security has layers — review code before you run it.
latestvk979znws8nyby2fh7daw6j8fbs84r5ne
License
MIT-0
Free to use, modify, and redistribute. No attribution required.