ClawHub
Back to skill

Security audit

科技新闻日报

Security checks across malware telemetry and agentic risk

Overview

This skill performs the stated news-report workflow, but it embeds and uses Feishu credentials and automatically grants document access through raw API calls.

Review before installing. This skill is coherent for generating and publishing a Feishu news digest, but the embedded Feishu app secret should be considered compromised and rotated. Install only after removing hard-coded credentials, replacing raw curl authentication with a managed Feishu integration, and adding explicit confirmation for document creation, permission grants, and sending messages to the fixed chat.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The fallback workflow directly retrieves a Feishu tenant access token using an app secret and then performs privileged API operations. Embedding or instructing use of application secrets in a skill that only needs to publish a daily news report creates unnecessary secret exposure and expands the blast radius if the skill, logs, or generated artifacts are accessed by others.

Missing User Warnings

High
Confidence
100% confidence
Finding
The skill exposes a hard-coded Feishu app secret in plain text and instructs using it to obtain an access token. Hard-coded credentials can be copied by anyone with access to the skill, reused outside the intended workflow, and abused to impersonate the application against Feishu APIs.

Missing User Warnings

High
Confidence
99% confidence
Finding
The fallback flow tells the agent to authenticate directly to Feishu Open API using an app secret, without any warning or secure secret-handling boundary. This encourages transmission and use of privileged credentials in ad hoc shell commands, increasing the chance of leakage through logs, transcripts, or operator reuse.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The reference content includes concrete Feishu API-style operations to create and write documents, including a folder token and owner identifier, without any warning, confirmation step, or discussion of data-modifying side effects. In the context of a skill that automatically generates and syncs daily reports to Feishu, this increases the risk of unintended document creation, overwriting content, or writing into a real user workspace without informed consent.

External Transmission

Medium
Category
Data Exfiltration
Content
--content @memory/YYYY-MM-DD-tech-news.xml

# 步骤 3：授予张公子 full_access 权限（用 Open API 直接调，绕过 lark-cli 交互限制）
ACCESS_TOKEN=$(curl -s -X POST 'https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal' \
  -H 'Content-Type: application/json' \
  -d '{"app_id": "cli_a94b4a1e43781cc7", "app_secret": "NALleHubGWREfZJkXpYizwszPIBFIBq7"}' \
  | python3 -c "import sys,json; print(json.load(sys.stdin).get('tenant_access_token',''))")
Confidence
97% confidence
Finding
curl -s -X POST 'https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal' \ -H 'Content-Type: application/json' \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
-d '{"app_id": "cli_a94b4a1e43781cc7", "app_secret": "NALleHubGWREfZJkXpYizwszPIBFIBq7"}' \
  | python3 -c "import sys,json; print(json.load(sys.stdin).get('tenant_access_token',''))")

curl -s -X POST 'https://open.feishu.cn/open-apis/drive/v1/permissions' \
  -H 'Authorization: Bearer '$ACCESS_TOKEN \
  -H 'Content-Type: application/json' \
  -d '{
Confidence
93% confidence
Finding
curl -s -X POST 'https://open.feishu.cn/open-apis/drive/v1/permissions' \ -H 'Authorization: Bearer '$ACCESS_TOKEN \ -H 'Content-Type: application/json' \ -d '{ "type": "docx", "token":

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.