Back to skill

Security audit

OfficeClaw

Security checks across malware telemetry and agentic risk

Overview

OfficeClaw is a clearly disclosed Microsoft Graph integration, with meaningful account-access risks that are mostly scoped by user-controlled permissions and safety gates.

Install in a dedicated environment, consider pinning officeclaw to a reviewed exact version, grant only the Microsoft Graph scopes you need, keep send/delete disabled unless required, and configure recipient and attachment-directory allowlists before allowing an agent to send mail or handle files.

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 (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:23
Finding
Unbounded Third-Party Package Installation Creates Supply-Chain Risk<![CDATA[ ## Vulnerability Details **File Location**: `SKILL.md`, lines 23–29 **Vulnerability Type**: Unbounded external executable dependency **Risk Level**: Medium ### Vulnerable Code ```bash pip install --upgrade "officeclaw>=1.1.1" ``` Or with uv: ```bash uv pip install --upgrade "officeclaw>=1.1.1" ``` ### Technical Analysis The installation instructions specify only a minimum package version. The `>=1.1.1` constraint permits package managers to install any later release published under the `officeclaw` package name. Consequently, the code installed and executed by users may differ from the version documented or reviewed with this Skill. The package implementation is not included in the audited project, which contains only `SKILL.md`. Therefore, the audit cannot verify the document's claims concerning OAuth token protection, recipient allowlists, attachment path validation, capability gates, or absence of telemetry. This is a supply-chain weakness rather than evidence that the currently published package is malicious. Exploitation would require compromise of the package publisher, package distribution infrastructure, or a future malicious release. ### Attack Path 1. An attacker compromises the package publisher account or otherwise causes a malicious future `officeclaw` release to be distributed from the configured Python package index. 2. A user follows the Skill instructions and runs `pip install --upgrade "officeclaw>=1.1.1"` or the equivalent `uv` command. 3. The resolver selects the malicious newer release because it satisfies the lower-bound-only version constraint. 4. Malicious code executes during package installation or when the user invokes the installed `officeclaw` executable. 5. The compromised package accesses resources available to the process, potentially including Microsoft Graph OAuth tokens, account data returned by Graph, email attachments, and locally readable files. ### Impact Assessment Successful exploitation could res ...[truncated 962 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Pin the dependency to a reviewed exact version, for example: ```bash pip install "officeclaw==1.1.1" ``` 2. Publish and verify cryptographic hashes using a requirements or lock file, and require hash checking during installation: ```bash pip install --require-hashes -r requirements.txt ``` 3. Configure installation to use an explicitly trusted package index and prevent unintended fallback to alternate indexes. 4. Vendor the reviewed implementation into the Skill package or include the corresponding source code so that capability gates, token handling, path restrictions, and network destinations can be audited. 5. Establish a controlled update process in which each new dependency version is reviewed, integrity-checked, and tested before the pin and hashes are updated. 6. Run the CLI with restricted operating-system privileges and narrowly scoped Microsoft Graph permissions. Keep send, delete, and attachment-download capabilities disabled unless required. 7. Use a dedicated virtual environment or sandbox and restrict filesystem and network access to the minimum resources needed for Microsoft Graph operations. ]]>
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.