Security Audit
LaunchDarkly Automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
LaunchDarkly Automation received a trust score of 89/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 2 findings: 0 critical, 0 high, 2 medium, and 0 low severity. Key findings include Skill enables powerful actions without least-privilege guidance, Unpinned external platform dependency (`rube` MCP).
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 17, 2026 (commit 99e2a295). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Skill enables powerful actions without least-privilege guidance The skill provides tools to create and delete LaunchDarkly flag triggers (`LAUNCH_DARKLY_CREATE_TRIGGER_WORKFLOW`, `LAUNCH_DARKLY_DELETE_TRIGGER_WORKFLOW`). These actions can significantly impact application behavior and stability by enabling or disabling features in production. The `SKILL.md` instructs users to "Connect your LaunchDarkly account when prompted (API key authentication)" but does not provide guidance on using an API key with the principle of least privilege (e.g., specifying minimum required scopes/roles). If a user provides an API key with broad administrative permissions, an attacker who compromises the agent or injects malicious prompts could perform highly destructive or disruptive actions via these tools. Add explicit guidance in the `SKILL.md` (or associated toolkit documentation) on the minimum required LaunchDarkly API key permissions (e.g., specific roles or custom roles) necessary for the skill's functionality. This encourages users to follow the principle of least privilege and limit the potential blast radius of a compromise. | LLM | SKILL.md:27 | |
| MEDIUM | Unpinned external platform dependency (`rube` MCP) The skill explicitly depends on an external platform, the Composio MCP server at `https://rube.app/mcp`, as indicated in the manifest (`"requires": {"mcp": ["rube"]}`) and the setup instructions. The security and integrity of this skill are directly tied to the trustworthiness and ongoing security of `rube.app`. There is no version pinning or specific integrity check mentioned for this external dependency. A compromise of `rube.app` could lead to malicious tool implementations being served, potentially misusing the provided LaunchDarkly API key or performing unauthorized actions. While direct version pinning for an external service might not be feasible, consider documenting the expected behavior and security posture of the `rube` MCP. For critical applications, evaluate the trust model for `rube.app` and consider alternatives or additional security controls. If possible, provide a mechanism for users to verify the integrity of the `rube` integration. | LLM | SKILL.md:1 |
Scan History
Embed Code
[](https://skillshield.io/report/91e247306720f0f8)
Powered by SkillShield