Back to skill

Security audit

企业微信对话配置

Security checks for vulnerabilities and agentic risk

Overview

This setup is not malicious, but it defaults to broad WeCom bot access while storing a bot secret locally and installing a mutable network-capable plugin.

Before installing, restrict WeCom access to known users or groups instead of using the open wildcard defaults, protect ~/.openclaw/openclaw.json with owner-only permissions, keep the bot secret out of repositories and shared backups, and prefer a pinned or otherwise verified plugin version when available.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Findings (3)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
SKILL.md:40
Finding

Unrestricted Direct-Message and Group Access

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:38
Finding

Bot Secret Stored in Plaintext Configuration

Content
View full analysis
", "secret": "", ``` ### Technical Analysis The setup directs users to place a reusable WeCom bot secret directly in `~/.openclaw/openclaw.json`. This stores the credential as plaintext in a regular configuration file. The document does not require restrictive file permissions or recommend a secret manager, environment-based reference, or operating-system credential store. Consequently, the secret may be exposed through permissive filesystem permissions, local account compromise, configuration backups, diagnostic bundles, accidental version-control commits, or support workflows. The placeholder itself is not a leaked credential. The vulnerability arises from the prescribed storage method after a user substitutes the real secret. ### Attack Path 1. An administrator replaces the placeholder with the production WeCom bot secret. 2. The secret is saved in plaintext in `~/.openclaw/openclaw.json`. 3. Another local user or process reads the file, or the file is copied into a backup, repository, support archive, or diagnostic package. 4. The attacker extracts the bot identifier and secret. 5. Subject to WeCom's server-side controls, the attacker reuses the credential to impersonate or establish an unauthorized connection for the bot. ### Impact Assessment Exposure of the credential may allow unauthorized use or impersonation of the WeCom integration. Potential consequences include: - Unauthorized connection attempts using the bot identity. - Interception or misuse of bot communications, depending on platform controls. - Abuse of the integration until the credential is revoked. - Expansion of impact if the same secret is reused elsewhere. The affected scope is primarily the configured bot and any OpenClaw or WeCom functionality re ...[truncated 42 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:21
Finding

Unpinned Installation of a Network-Capable Plugin

Content
View full analysis
Remediation
View remediation
``` 2. Verify the package publisher, registry source, signature, and integrity checksum before installation. 3. Use a lockfile or equivalent immutable dependency record where supported. 4. Review release notes and source changes before upgrading the pinned version. 5. Run the OpenClaw process under a dedicated, low-privilege operating-system account. 6. Limit the plugin's filesystem, environment-variable, and network access to the minimum required endpoints and data. 7. Do not normalize or dismiss the security warning; explain the required permissions and require administrators to verify them. 8. Maintain a tested rollback procedure and rotate credentials if a dependency compromise is suspected. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill directs the user to store a bot secret in plaintext inside a local JSON configuration file but does not warn that this credential is sensitive, should be protected with restrictive file permissions, and should not be committed, shared, or logged. While local config storage may be functionally required by the platform, the absence of handling guidance increases the risk of accidental credential disclosure through backups, screenshots, repo commits, or multi-user systems.

Content

No source excerpt is available for this finding.