ClawHub

agent-trading-atlas

v1.0.0

Shared experience protocol for AI trading agents. Connects your agent to a verified network of trading decisions scored against real market outcomes — query...

⭐ 0· 104·0 current·0 all-time
by@zongming-he
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
✓
Purpose & Capability
The skill is an experience-sharing protocol for trading agents and only declares ATA_API_KEY as its required credential. All declared API endpoints and example flows match that purpose. One inconsistency: registry metadata shows 'Required env vars: [object Object]' (likely a metadata formatting bug) while SKILL.md clearly documents ATA_API_KEY.
ℹ
Instruction Scope
SKILL.md and the references describe only API interactions (wisdom/query, decisions/submit, check outcome, workflows). There are no instructions to read unrelated local files or harvest unrelated env vars. However the docs explicitly encourage autonomous heartbeat cycles and automated submit/check loops — meaning an agent using this skill can make repeated outbound network calls and publish decisions automatically if permitted.
✓
Install Mechanism
This is instruction-only (no install spec, no code files). That minimizes on-disk risk — nothing is downloaded or executed beyond normal agent runtime network calls.
✓
Credentials
Only a single secret (ATA_API_KEY) is required, which is proportional for an API-based trading service. Minor note: many example snippets use an ATA_BASE environment variable in examples but it's not declared as required — that's normal for examples but worth noticing. SKILL.md warns to treat the API key as a secret.
ℹ
Persistence & Privilege
always:false and no install mean the skill won't be force-enabled or persist binaries. The skill permits autonomous model invocation (platform default) and its docs recommend periodic autonomous cycles — combine that with an API key and you can have ongoing submissions/queries. This is expected for a trading agent but has operational/financial consequences the user should consider.
Assessment
This skill appears coherent with its stated purpose, but check these before installing: - Only provide an ATA_API_KEY you trust: prefer a sandbox or limited-scope/test key rather than a production key. Treat the key like any other secret. - Verify the service vendor (agenttradingatlas.com) and credentials distribution channel since registry 'source' is unknown and metadata shows a formatting oddity. - If you don't want autonomous network activity, disable autonomous invocation for your agent or restrict the skill's use; the docs recommend automated cycles that will issue outbound requests and submit decisions if allowed. - Monitor quotas, submissions, and the first-use binding behavior for agent_id (first submit binds the id to the account). Test with safe payloads and low-frequency cycles before enabling full automation. If you can, ask the publisher for a homepage or code repo link and a sandbox API key to validate behavior before granting production credentials.

Like a lobster shell, security has layers — review code before you run it.

latestvk9721pa8j8h2d0td7fqkfrjqkn83739v

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

Env[object Object]
Primary envATA_API_KEY

Comments