Clawhub Publisher

Security checks across static analysis, malware telemetry, and agentic risk

Overview

The skill has a plausible publishing purpose, but it asks users to rely on unreviewed external packages and ClawHub API-key powers that can publish, roll back, and change team access.

Review before installing. Only use this if you trust the external npm/PyPI package and repository, and prefer a least-privilege ClawHub token. Start with validation or dry-run mode, avoid global installs and automatic CI deployment until verified, and require explicit approval for publish, rollback, batch publish, and team-access changes.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

A user could end up running unreviewed external code while giving it publishing credentials and authority over ClawHub releases.

Why it was flagged

The skill tells users or agents to fetch executable code from external package registries, but the submitted ClawHub artifact has no install spec and no implementation code to review; package.json references index.js and bin/cli.js that are not present in the provided files.

Skill content
npm install clawhub-publisher
# or
python -m pip install clawhub-publisher
...
npm install -g clawhub-publisher
Recommendation

Verify the npm/PyPI package ownership and source repository, inspect the external package before use, pin versions, and avoid global or CI installation until the implementation is trusted.

#
ASI03: Identity and Privilege Abuse
Medium
What this means

If a broad API key is used, the tool or an agent invoking it could modify public skill releases or collaborator permissions beyond what the user intended.

Why it was flagged

The documented workflows use ClawHub credentials and can publish, roll back, and grant team access, but the registry metadata declares no required environment variables or primary credential and the docs do not describe least-privilege scopes.

Skill content
apiKey: process.env.CLAWHUB_API_KEY ... clawhub-publisher login --token YOUR_API_KEY ... await publisher.rollback(...) ... await publisher.addTeamMember({ ... role: 'publisher' })
Recommendation

Use a least-privilege ClawHub token if available, avoid giving team-management or rollback permissions unless needed, and require explicit approval before publish, rollback, or access-change operations.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

A mistaken path, prompt, or CI trigger could publish multiple skills or make unwanted release changes.

Why it was flagged

Batch publishing and direct publishing are expected for a deployment tool, but the examples show broad mutation capabilities and a non-dry-run default.

Skill content
clawhub-publisher publish-batch --dir ./skills
...
dryRun: false  // Preview changes without publishing
Recommendation

Run validation and dry-run mode first, use explicit skill paths, and review generated changelogs, README changes, and target versions before publishing.