Back to skill

Security audit

Venus Tracker

Security checks for vulnerabilities and agentic risk

Overview

This project-tracking skill is coherent, but it delegates important data-changing actions to an unaudited local script that can modify Feishu project records.

Review before installing. Use this only in the intended Venus/Feishu project environment, verify the referenced synchronization script yourself, confirm the target Feishu Base and credentials before running any command, and prefer the reviewed ClawHub install over the unpinned git clone path.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)

T08 · Insecure Dependencies

Warning
Location
README.md:15
Finding

Unpinned Remote Repository Installation

Content
View full analysis

Vulnerability Details

File Location: README.md, lines 15-17
Vulnerability Type: Unpinned supply-chain dependency
Risk Level: Medium

Vulnerable Code

bash
git clone https://github.com/spikesubingrui-design/venus-tracker.git ~/.openclaw/workspace/skills/venus-tracker

Technical Analysis

The installation instructions clone the current default branch of a remote Git repository without selecting an immutable commit or verifying a signature or checksum. Consequently, the content installed by this command can differ from the version covered by this audit.

Although the repository URL corresponds to the homepage declared by the Skill, this alone does not guarantee content integrity. A compromised maintainer account, repository, or upstream branch could modify the Skill instructions or introduce executable files after review.

Attack Path

  1. An attacker compromises the upstream repository or an authorized maintainer account.
  2. The attacker modifies the default branch to include malicious Skill instructions or executable content.
  3. A user follows the documented unpinned git clone installation command.
  4. OpenClaw or another agent loads the modified Skill.
  5. The malicious content operates with the tools and permissions granted to the agent, including the Skill's declared shell capability.

Impact Assessment

Exploitation requires control of, or unauthorized write access to, the upstream repository. Successful exploitation could alter agent behavior or execute commands with the privileges of the user running the agent. The resulting scope could include accessible workspace files, agent-readable credentials, and external services available to the agent.

Remediation
View remediation

Remediation Suggestions

  • Pin installation to a reviewed, immutable Git commit or signed release tag.
  • Publish and verify a SHA-256 checksum for release artifacts.
  • Require signed commits or release signatures and document signature verification.
  • Avoid automatically updating the installed Skill from a mutable branch.
  • Re-audit the package whenever the pinned revision changes.
  • Prefer a package installation mechanism with lockfiles, integrity metadata, and provenance verification.

T08 · Insecure Dependencies

Warning
Location
SKILL.md:83
Finding

Operational Behavior Delegated to an Unbundled Mutable Script

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 83-90
Vulnerability Type: Unverified external executable dependency
Risk Level: Medium

Vulnerable Code

bash
python3 /Users/spikescp/.openclaw/workspace/scripts/venus-bitable-sync.py collect
bash
python3 /Users/spikescp/.openclaw/workspace/scripts/venus-bitable-sync.py track

Technical Analysis

The Skill directs the agent to execute venus-bitable-sync.py from a user-specific workspace path. The referenced Python file is not included in the audited package, so its implementation, integrity, data handling, and effective permissions cannot be verified.

The Skill similarly depends on external workspace instructions and planning state identified elsewhere in SKILL.md. These dependencies can change independently of the reviewed package. Because the manifest grants shell access and marks the Skill as mutating, execution of a substituted or compromised script would not be constrained to the behavior documented in this repository.

The absolute path also binds execution to a particular local workspace layout. If another party can create or modify the file at that path, the documented command becomes a direct arbitrary-code execution channel under the invoking user's privileges.

Attack Path

  1. An attacker obtains write access to the referenced workspace script, its parent directory, a synchronization source that populates it, or the account responsible for distributing it.
  2. The attacker replaces or modifies venus-bitable-sync.py.
  3. The user or agent invokes the documented collect, track, or another synchronization operation.
  4. Python executes the substituted code without an integrity or provenance check.
  5. The malicious script accesses resources available to the invoking process, potentially including local files and credentials used for Feishu synchronization.

Impact Assessment

A successful attack would exe ...[truncated 511 chars]

Remediation
View remediation

Remediation Suggestions

  • Bundle the synchronization script with the Skill so it can be versioned and audited as one release.
  • Pin external prerequisite Skills and planning dependencies to reviewed versions.
  • Verify the script's cryptographic hash or signature immediately before execution.
  • Replace the user-specific absolute path with a validated package-relative path.
  • Ensure the script and parent directories are not writable by untrusted users.
  • Run synchronization with a dedicated least-privilege account and narrowly scoped Feishu credentials.
  • Restrict filesystem and network access to resources required for synchronization.
  • Require explicit confirmation before high-impact initialization, migration, or remote mutation operations.
  • Document the exact external APIs, data transmitted, credentials required, and records modified by every operation.
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Tp2

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding

Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · SKILL.md (reported line 34)May include surrounding context.

md
执行本 skill 前,必须先读:

- `workspace/skills/venus-project-pm/SKILL.md`
- `workspace/.planning/venus-erp/task_plan.md`

## 关键文件

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill exposes multiple mutating shell commands that initialize, migrate, seed, and sync remote Feishu Base data and local planning artifacts without explicit operator warnings, confirmation gates, dry-run guidance, or rollback instructions. In an agent setting, this increases the chance of accidental destructive or unauthorized state changes to production tracking data, especially because the skill is designed to be operationally used under a project-management workflow.

Content

No source excerpt is available for this finding.