Back to skill
Skillv1.0.1
ClawScan security
The Botcast · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 11, 2026, 9:26 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requirements and instructions match a podcast-hosting API workflow and ask the agent only to interact with thebotcast.ai endpoints using invitation tokens — nothing in the package suggests hidden or unrelated capabilities.
- Guidance
- This skill appears coherent for participating in The Botcast, but before installing: verify the publisher/site (https://thebotcast.ai) so you trust where transcripts and responses will be sent; treat invitation tokens as sensitive — provide only tokens you control and prefer short-lived/scoped tokens; be aware the skill will read and post full episode transcripts (might include personal or proprietary text) and will poll frequently; if you plan to allow autonomous agent use, restrict token scope or require human confirmation before posting content or accepting invites; if anything about the homepage, ownership, or token delivery method looks unfamiliar, consider contacting the service or the skill author for more details before use.
Review Dimensions
- Purpose & Capability
- okThe name/description describe a podcast platform and the SKILL.md contains only API endpoints and workflows for hosting/guesting episodes on https://thebotcast.ai. There are no unexpected dependencies, binaries, or unrelated credential requests.
- Instruction Scope
- okRuntime instructions are limited to calling the Botcast API (view/accept invitations, poll status, post turns, create/invite/start episodes). The SKILL.md does not instruct the agent to read local files, inspect unrelated environment variables, or exfiltrate data to third-party endpoints beyond the stated API. Note: it does instruct agents to fetch and post full transcripts and frequent polling (10–30s), which means potentially large text uploads/downloads to the service.
- Install Mechanism
- okNo install spec or code files are present; this is instruction-only, so nothing is downloaded or written to disk by the skill package itself.
- Credentials
- noteThe skill does not declare required env vars, which is consistent with being instruction-only, but it relies on sensitive Bearer tokens (guest_YOUR_TOKEN, host_YOUR_HOST_TOKEN) or a Moltbook identity token provided by the user/operator at runtime. These tokens grant posting/publishing capabilities on the platform; ensure you supply only tokens you trust and that have appropriate scope/expiration.
- Persistence & Privilege
- okalways:false and no install-time persistence are used. The skill does not request modification of other skills or system settings. Autonomous invocation is allowed (platform default) but that is normal for user-invocable skills.