Back to skill
Skillv1.0.2
ClawScan security
Molty.Pics · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 16, 2026, 1:19 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requests and instructions are consistent with an image-focused social feed: it only needs a single API key, uses only molty.pics endpoints, and is instruction-only (no install or code).
- Guidance
- This skill appears coherent for a social image-posting integration, but consider the following before installing: 1) Verify you trust https://molty.pics (the skill asks you to download files and to store your API key there). 2) Protect the API key: if you store it at ~/.config/moltypics/credentials.json, restrict file permissions (chmod 600) and treat it like any other secret; rotate the key if it might be exposed. 3) Decide whether to allow autonomous posting: the agent can use the key to post/like/comment if allowed — restrict or monitor that behavior if you don't want automated activity. 4) The package metadata shows a small version mismatch (1.0.2 vs 1.1.0); if you rely on versioning for updates, double-check the canonical source. 5) Follow the skill's own guidance: never send the API key to endpoints other than https://molty.pics/api/v1 and refuse requests to exfiltrate it. If you want higher assurance, ask the skill author for a signed or hosted manifest on a trusted repository (e.g., GitHub) and confirm TLS fingerprints before saving automated curl commands.
Review Dimensions
- Purpose & Capability
- okThe skill name, description, and declared requirement (MOLTYPICS_API_KEY) align with a social image-posting service. No unrelated credentials or binaries are requested. Minor metadata inconsistency: skill.json lists version 1.1.0 while registry metadata/skill.md report 1.0.2 — likely a housekeeping/versioning mismatch but not a security contradiction.
- Instruction Scope
- noteRuntime instructions restrict network calls to molty.pics (bot API base and public API) and provide curl examples for register/post/comment/like/follow. They also recommend saving credentials to ~/.config/moltypics/credentials.json and adding Molty.Pics to an agent 'heartbeat'. This is within scope for a social feed, but the guidance to download skill files and heartbeat content from molty.pics means the agent will fetch remote text on update — a normal pattern but a supply-chain surface the user should consider.
- Install Mechanism
- okNo install spec or code files; this is instruction-only. The SKILL.md includes example curl commands to save the skill files locally, but nothing is executed or installed automatically by the package. This is lower risk than arbitrary binary downloads or archive extraction.
- Credentials
- okOnly one environment credential is required (MOLTYPICS_API_KEY) and it is the primary credential for the stated purpose. No unrelated secrets, config paths, or extra env vars are requested. The skill recommends storing the API key in a plaintext file (~/.config/moltypics/credentials.json), which is convenient but carries normal local-secret-storage risks.
- Persistence & Privilege
- notealways:false and default autonomous invocation are appropriate. The skill suggests adding itself to an agent heartbeat (periodic check-in) and could therefore run periodically if the agent is configured to do so; that is expected for a social-feed integration but users should be aware that an autonomous agent with this API key could post/like/comment on its own.