v1.1.0

FHE-as-a-Service

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 8:00 AM.

Analysis

This skill asks agents to send clinical values to a remote API while making strong privacy and compliance claims that are not supported by its own usage examples.

GuidanceReview carefully before installing or using this skill. Do not send real patient data unless you have verified how client-side FHE encryption works, what exact data is transmitted, and whether the provider's HIPAA/GDPR/no-PHI-logging claims are backed by appropriate documentation.

Findings (3)

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Human-Agent Trust Exploitation

SeverityHighConfidenceHighStatusConcern

SKILL.md

"The server never sees plaintext patient data" ... "curl -X POST https://rheumascore.xyz/fhe/v1/compute/das28" ... "-d '{"values": [10, 5, 40, 60]}'"

The artifact makes strong privacy claims, but its own compute example sends raw-looking clinical values to the remote service with no documented client-side encryption step.

User impactA user or agent may believe patient data is protected by FHE and send sensitive clinical information to a third-party service in plaintext or in an unclear format.

RecommendationDo not use real patient data unless the provider documents and demonstrates client-side encryption, ciphertext input schemas, and independently verifiable privacy/compliance controls.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

SKILL.md

"Response includes your `api_key`" ... "Authorization: Bearer fhe_<your_key>"

The skill expects agents to obtain and use a bearer API key, while the registry metadata declares no primary credential or required environment variables.

User impactThe API key should be treated as a secret and may enable use of the service, including paid usage after the free tier.

RecommendationStore the API key securely, avoid exposing it in prompts or logs, and clarify the credential requirement in the metadata.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityHighConfidenceHighStatusConcern

SKILL.md

"Base URL: `https://rheumascore.xyz/fhe/v1`" ... "Compute a Score (FHE-encrypted)" ... "-d '{"values": [10, 5, 40, 60]}'"

The skill routes clinical-score inputs to an external provider API, but the documented data boundary does not clearly show that the transmitted values are ciphertext rather than sensitive patient measurements.

User impactSensitive clinical inputs could be disclosed to a remote provider under unclear privacy guarantees.

RecommendationRequire explicit user approval before sending any clinical data, confirm the exact data sent, and avoid protected health information unless encryption and handling guarantees are verified.