Trust Assessment
langfuse received a trust score of 75/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 3 findings: 0 critical, 1 high, 2 medium, and 0 low severity. Key findings include LLM Interaction Data Sent to External Observability Service, Use of Sensitive API Keys with Configurable Endpoint, Unpinned External Dependencies.
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 20, 2026 (commit e36d6fd3). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Use of Sensitive API Keys with Configurable Endpoint The skill requires and uses a `secret_key` (and `public_key`) to authenticate with the Langfuse service. The `host` URL for Langfuse is also configurable. If a malicious `host` is provided, the `secret_key` could be transmitted to an attacker-controlled server, leading to credential harvesting. The examples show these keys hardcoded as placeholders, which is a bad practice for actual deployment. Never hardcode API keys directly in code. Use secure methods for managing credentials, such as environment variables, a dedicated secret management service, or a secure configuration file. Validate the `host` URL to ensure it points to a legitimate Langfuse instance. | LLM | SKILL.md:21 | |
| MEDIUM | LLM Interaction Data Sent to External Observability Service The skill's primary function is to send sensitive LLM interaction data (prompts, responses, user IDs, session IDs, metadata) to an external Langfuse observability service. This data is transmitted to a configurable `host` using `public_key` and `secret_key` for authentication. While this is the intended purpose of Langfuse, it represents a significant data outflow that could be exploited if the `host` or API keys are compromised, or if the data itself is highly sensitive and not intended for third-party processing. Ensure that the Langfuse `host` is trusted and that API keys (`public_key`, `secret_key`) are securely managed (e.g., via environment variables, secret management systems) and have appropriate access controls. Review the type of data being sent to Langfuse to ensure compliance with data privacy policies. | LLM | SKILL.md:20 | |
| MEDIUM | Unpinned External Dependencies The skill implicitly relies on external Python libraries such as `langfuse`, `langchain`, and `openai`. The provided `SKILL.md` does not specify exact version numbers for these dependencies. This lack of pinned versions introduces a supply chain risk, as future updates to these libraries could introduce breaking changes, vulnerabilities, or unexpected behavior without explicit review. For production deployments, always pin exact versions of all external dependencies in a `requirements.txt` or similar dependency management file. Regularly review and update dependencies to incorporate security patches while mitigating the risk of unexpected changes. | LLM | SKILL.md:1 |
Scan History
Embed Code
[](https://skillshield.io/report/190735197b03b5c0)
Powered by SkillShield