Back to skill

Security audit

Gog2

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a legitimate Google Workspace CLI helper, but it needs Review because it enables broad Google account access and write/export actions with limited safety scoping.

Install only if you trust the `gog` CLI source and are comfortable granting Google Workspace access. Prefer a dedicated or least-privilege Google account, grant only needed services where possible, keep the OAuth client secret in a secure local path, verify the Homebrew package provenance before authenticating, and require explicit confirmation before sending email, changing Sheets, clearing ranges, copying/exporting Docs, or writing calendar data.

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:5
Finding
Unpinned CLI Installation from a Third-Party Homebrew Tap## Vulnerability Details **File Location**: `SKILL.md`, line 5 **Vulnerability Type**: Third-party supply-chain dependency installed without version or integrity pinning **Risk Level**: Medium ### Vulnerable Code Snippet ```yaml metadata: {"clawdbot":{"emoji":"🎮","requires":{"bins":["gog"]},"install":[{"id":"brew","kind":"brew","formula":"steipete/tap/gogcli","bins":["gog"],"label":"Install gog (brew)"}]}} ``` ### Technical Analysis The skill instructs the platform to install `gog` from the third-party Homebrew tap `steipete/tap/gogcli`. The dependency declaration does not identify an immutable release, commit, package digest, or expected checksum. Consequently, the code installed at a later date may differ from the dependency that was available when this skill was reviewed. If the tap, formula repository, maintainer account, release infrastructure, or referenced package artifact is compromised, an attacker could replace the expected CLI with a malicious build. Homebrew formula installation may also execute formula-defined build or installation logic under the invoking user's account. This finding does not establish that the current `gog` package or its maintainer is malicious. The risk arises from reliance on mutable third-party supply-chain state without integrity verification. ### Attack Path 1. An attacker compromises the third-party tap, its maintainer account, release infrastructure, or an artifact referenced by the formula. 2. The attacker changes the formula or distributed artifact while preserving the expected package name. 3. A user or agent installs the dependency through the skill's declared Homebrew installation method. 4. Homebrew retrieves and executes or installs the attacker-controlled package content. 5. The malicious package executes with the privileges of the user running Homebrew. 6. The substituted `gog` binary can later intercept OAuth setup, tokens, command arguments, and Google Workspace data ...[truncated 897 chars]
Remediation
## Remediation Suggestions 1. Pin the dependency to a specifically reviewed release rather than relying solely on a mutable formula name. 2. Verify downloaded artifacts against a publisher-provided cryptographic checksum or signature before installation. 3. Where supported, reference an immutable source commit or versioned artifact and record its expected digest in the installation metadata. 4. Prefer an official, trusted distribution channel with signed releases and documented provenance. 5. Add installation documentation that identifies the expected package version, checksum, signer, and verification procedure. 6. Use automated dependency monitoring to detect upstream ownership changes, formula modifications, checksum changes, and newly disclosed vulnerabilities. 7. Perform OAuth setup only after verifying the installed binary's provenance and integrity. 8. Grant only the Google Workspace scopes required for the intended operation, separating accounts or authorizations when broad multi-service access is unnecessary.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Credential Access

High
Category
Privilege Escalation
Content
Use `gog` for Gmail/Calendar/Drive/Contacts/Sheets/Docs. Requires OAuth setup.

Setup (once)
- `gog auth credentials /path/to/client_secret.json`
- `gog auth add you@gmail.com --services gmail,calendar,drive,contacts,sheets,docs`
- `gog auth list`
Confidence
70% confidence
Finding
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill instructs users to provide an OAuth client secret file but gives no warning that this file contains sensitive credentials and must be stored, transmitted, and permissioned securely. In an agent context, users may paste local paths or expose secrets in logs, prompts, shared environments, or shell history, increasing the chance of credential leakage and unauthorized API access.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill lists commands that can update, append, clear, copy, export, and send data across Google Workspace without consistent safety warnings or confirmation requirements. In an agent-operated environment, these examples normalize write and destructive actions and can lead to accidental data modification, deletion, exfiltration, or unintended communications if invoked from loosely scoped user requests.

Static analysis

No suspicious patterns detected.