Skillv1.1.1

ClawScan security

web123 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 18, 2026, 5:01 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only catalog/recommender for AntalphaAI Web3 skills that reads a local skills.json and emits install recommendations; it requests no credentials, does not install code itself, and its requirements align with its stated purpose.
Guidance: This skill is a simple local catalog/recommender for AntalphaAI Web3 skills and appears internally consistent. Before installing or acting on its recommendations: 1) Verify the GitHub URLs it lists (github.com/AntalphaAI/...) are the official repositories you expect, since registry metadata shows Source: unknown. 2) Inspect any recommended skill's repo and its install/requirements before running openclaw skill install — some recommended skills (e.g., cex-trader) will request API keys or other permissions. 3) Avoid blind batch-installing many skills; install only the subset you trust and have reviewed. 4) If you need higher assurance, review the referenced repositories' code or their maintainers' identities before enabling them in your agent.

Review Dimensions

Purpose & Capability: noteThe skill is a navigator/recommender for AntalphaAI skills and only needs a local metadata file (references/skills.json) to operate. It does not request binaries, env vars, or credentials, which is consistent. Minor note: registry metadata lists Source: unknown and no homepage, while SKILL.md and README point to github.com/AntalphaAI — you should verify those GitHub repos are the expected official sources.
Instruction Scope: okSKILL.md directs the agent to load the local references/skills.json, perform simple token/keyword matching, and output templated recommendations and install commands. It does not instruct reading unrelated system files, accessing environment variables, or sending data to external endpoints. It does create install commands that, if executed, will fetch other skills from GitHub (expected behavior for a navigator).
Install Mechanism: okThis is instruction-only with no install spec and no code files that get written/executed by the skill itself. That is the lowest-risk install model. The README and templates reference 'openclaw skill install <github-url>' which delegates installation to the platform — normal for a catalog skill.
Credentials: okThe skill declares no required environment variables, no primary credential, and no config paths. That matches its role as a read-only recommender. Note that some skills it recommends (e.g., cex-trader) will legitimately require API keys; those are external to this skill and should be reviewed individually before installation.
Persistence & Privilege: okalways is false and the skill is user-invocable (normal). It does not request persistent system-level privileges or modify other skills' configurations. Autonomous invocation remains allowed (platform default) but is not combined with any broad credential access here.