Back to skill

Security audit

Brigade Handoffs

Security checks for vulnerabilities and agentic risk

Overview

This skill gives clear local-first guidance for using Brigade handoffs, with the main caution that it installs an unpinned third-party CLI.

Before installing, consider pinning or verifying the Brigade CLI package source and version. Review the dry-run output before allowing setup changes, and do not run the install or Brigade commands with elevated privileges unless you have a separate reason to do so.

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:25
Finding
Unpinned Third-Party CLI Installation## Vulnerability Details **File Location**: `SKILL.md`, lines 25 and 35 **Vulnerability Type**: Unpinned third-party dependency installation **Risk Level**: Medium ### Vulnerable Code The same installation command appears in both setup workflows: ```bash pipx install brigade-cli ``` ### Technical Analysis The skill instructs users to install `brigade-cli` from the package index without pinning an exact reviewed version or verifying a cryptographic hash or signature. As a result, the installed artifact may change after the skill itself has been reviewed. The instructions subsequently execute numerous `brigade` commands against the user's repository or agent workspace. Python package installation and CLI execution can run package-controlled code with the privileges of the invoking user. If the package, one of its transitive dependencies, its publisher account, or the configured package index is compromised, following these instructions could execute attacker-controlled code. The audit did not find evidence that `brigade-cli` is currently malicious. The finding concerns the unsafe and mutable trust boundary created by the unpinned installation. ### Attack Path 1. An attacker compromises the `brigade-cli` publishing account, a transitive dependency, or the package distribution channel. 2. The attacker publishes a malicious release under the expected package name. 3. A user follows the skill and runs `pipx install brigade-cli`. 4. `pipx` retrieves the current attacker-controlled release because no exact version or artifact hash is specified. 5. Malicious code executes during installation or when a subsequent `brigade` command is invoked. 6. The payload accesses or modifies data available to the user's account, potentially including the target workspace and locally available credentials. ### Impact Assessment Successful exploitation could provide arbitrary code execution with the privileges of the user running `pipx` or `brigade`. The affected scope may in ...[truncated 611 chars]
Remediation
## Remediation Suggestions 1. Replace the mutable installation command with an exact, reviewed version: ```bash pipx install "brigade-cli==X.Y.Z" ``` 2. Document the authoritative package source and upstream repository so users can validate package provenance. 3. Install from a trusted, explicitly configured package index rather than relying implicitly on the user's package-manager configuration. 4. Where supported, verify the downloaded artifact using a published cryptographic hash or trusted signature before installation. 5. Review and lock relevant transitive dependencies as part of each approved release. 6. Retain `pipx` isolation and avoid running installation or CLI commands with `sudo` or another privileged account. 7. Test upgrades separately, review release changes, and update the pinned version only after validation.
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

Static analysis

No suspicious patterns detected.