Back to skill
Skillv1.0.0
ClawScan security
Clarity Annotate · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousFeb 25, 2026, 8:32 PM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill appears to implement the stated Clarity Protocol annotation functionality, but there are mismatches between the declared registry metadata and the runtime instructions (missing required env var and undeclared dependency), which warrants caution before installing.
- Guidance
- This skill's code and documentation implement the advertised annotation functionality and will send data to https://clarityprotocol.io. Before installing: (1) Confirm you are comfortable giving a CLARITY_WRITE_API_KEY (write capability) — the SKILL.md requires it but the registry metadata does not advertise it; do not supply production or sensitive credentials without verifying the service and key scope. (2) Verify the clarityprotocol.io homepage and that the API endpoint is legitimate. (3) Ensure the Python 'requests' package is available on the agent runtime (the skill has no install step). (4) Understand that annotations (content you submit) will be transmitted to an external service — avoid including sensitive patient data or secrets in annotation content. (5) Prefer obtaining the write API key with minimal privileges and confirm rate limits and error handling fit your operational needs. The inconsistencies in declared requirements are likely an oversight but merit caution; resolve them (ask the publisher to declare required env vars and dependencies) before trusting the skill with real data.
Review Dimensions
- Purpose & Capability
- noteThe name, description, SKILL.md, and code files all align: the scripts call https://clarityprotocol.io/api/v1 to list and submit annotations. Requesting an API key for write operations (CLARITY_WRITE_API_KEY) is appropriate for this purpose. However, the registry metadata lists no required environment variables while SKILL.md and the code clearly require CLARITY_WRITE_API_KEY (and optionally CLARITY_API_KEY) — an inconsistency that should be resolved.
- Instruction Scope
- okSKILL.md and the scripts are narrowly scoped to calling the Clarity API to list/submit annotations. The runtime instructions and code only perform network calls to the declared API_BASE and do not read arbitrary local files or unrelated environment variables. They do exit on API errors and print error details to stderr, which is expected behavior.
- Install Mechanism
- noteThere is no install spec (instruction-only), which minimizes install-time risk. However, the code depends on the third-party Python package 'requests' but the package/dependency is not declared in registry metadata or installation instructions — this is an operational omission (the scripts will fail if 'requests' is not present).
- Credentials
- noteThe only credentials used are CLARITY_WRITE_API_KEY for write operations and optionally CLARITY_API_KEY for read operations, which are proportionate to the stated functionality. The problem is that the registry metadata does not advertise these required env vars, so a user or automated permission check may not realize a write-capable secret is needed.
- Persistence & Privilege
- okThe skill does not request persistent or elevated privileges and is not marked always:true. It does not modify other skills or system-wide settings. Execution simply runs the provided scripts which perform API calls.