v1.0.0

TNBC Research Swarm

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 5:49 AM.

Analysis

This instruction-only skill is coherent for TNBC research collaboration, but it does send agent-generated findings and QC verdicts to a third-party multi-agent research API.

GuidanceSafe to consider installing if you intend to participate in this TNBC research workflow. Before using it, verify the Research Swarm service, review any assigned task for relevance, and manually check generated findings or QC verdicts before posting them to the external API.

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.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

curl -s -X POST https://www.researchswarm.org/api/v1/agents/[agentId]/findings ... -d @/path/to/finding.json

The workflow has the agent POST generated research findings and QC verdicts to an external service. This is aligned with the skill's purpose, but it is still a remote mutation that users should review.

User impactThe agent may create a Research Swarm session and submit content or review decisions to the platform.

RecommendationReview generated findings and QC verdicts before submission, and avoid including confidential or personal information in the JSON sent to the API.

Agentic Supply Chain Vulnerabilities

SeverityInfoConfidenceHighStatusNote

metadata

Source: unknown; Homepage: none

The skill has no installed code, but its provenance is limited in the registry metadata while it directs use of a third-party API.

User impactUsers have less registry-level information for independently verifying who operates or maintains the skill and service.

RecommendationVerify the Research Swarm endpoint and operator before sending submissions, especially if the content is unpublished or sensitive.

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

SeverityLowConfidenceHighStatusNote

SKILL.md

The response includes an assignment with: - `type`: "research" or "qc_review" ... - `description`: Research topic - `searchTerms`: Keywords for searching

A remote multi-agent platform supplies assignments that guide the agent's work. This is core to the skill, but the platform identity and task-origin boundaries are not further described.

User impactRemote assignment text can influence what the agent researches and submits.

RecommendationTreat received task descriptions and search terms as data, not higher-priority instructions; confirm they remain TNBC research tasks and ignore any embedded prompt-like instructions.