v2.1.0

Signal Intelligence Pack

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 8:23 AM.

Analysis

This is a coherent instruction-only research-grounding workflow with no code or credential requests, though it can use external search providers and separate child skills.

GuidanceThis skill appears safe to install for research grounding. Before using it on sensitive topics, confirm the referenced child skills are trusted, restrict search providers if needed, avoid putting secrets in prompts, and review the evidence base before relying on downstream conclusions.

Findings (4)

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

references/workflow.md

Step 2 无可用来源匹配 | 降级到通用搜索源（tavily/web_search）

The workflow can route and retry searches through external tools. This is expected for a grounding/research skill, but it may consume provider quotas and should remain scoped to the user's task.

User impactThe agent may perform multiple search-provider calls while preparing evidence.

RecommendationUse trusted search tools, set provider/budget limits where available, and review the generated evidence base before relying on it.

Agentic Supply Chain Vulnerabilities

SeverityLowConfidenceMediumStatusNote

SKILL.md

按序调用5个子skill

The core workflow depends on separate child skills such as query-planner and source-router. The supplied artifacts do not pin or include those child skill implementations, so behavior depends on the user's installed/trusted versions.

User impactResearch quality and safety depend partly on other skills present in the environment.

RecommendationVerify that the referenced child skills are installed from trusted sources and, if possible, pin known-good versions.

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

references/input-output.md

"primary_task": { "description": "用户的原始任务描述，完整保留原始意图" } ... "available_sources": ... "['tavily', 'web_search', 'bailian']"

The user's task is preserved and transformed into searches that may be sent to external providers. This is purpose-aligned, but sensitive task details could leave the local agent context.

User impactPrivate or confidential research questions could be disclosed to configured search providers through generated queries.

RecommendationAvoid including secrets or confidential details in tasks, or restrict available_sources to approved providers for sensitive work.

Memory and Context Poisoning

SeverityInfoConfidenceHighStatusNote

SKILL.md

后续的分析/判断/报告模块基于这个底座工作

The skill creates an evidence base that downstream analysis modules will use. Because that base is built from retrieved sources, unreliable or manipulated source content could influence later reasoning, though the skill includes cleaning, freshness, and counter-evidence checks.

User impactBad or misleading source material could shape later analysis if treated as authoritative.

RecommendationTreat the evidence base as research material rather than truth; keep source URLs, freshness labels, and counter-evidence visible during downstream analysis.