Trust Assessment
tensorpm received a trust score of 82/100, 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, 1 high, 1 medium, and 0 low severity. Key findings include Unauthenticated Local A2A API, Potential Arbitrary File Read via `documentPath`.
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 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Unauthenticated Local A2A API The TensorPM A2A agent endpoint runs on `localhost:37850` and, by default, explicitly states 'No authentication required'. This means any process running on the user's local machine can access, create, modify, and delete project data, action items, and other sensitive information within TensorPM without any authorization checks. This poses a significant risk for data exfiltration or manipulation if another local application or malware compromises the user's system, as it grants full control over TensorPM's data to any local actor. Strongly recommend enabling authentication by default for the A2A endpoint. If an `A2A_HTTP_AUTH_TOKEN` is available, make its configuration mandatory for any write operations or sensitive data access. Provide clear instructions for users to configure this token and emphasize the security implications of running without it. | LLM | SKILL.md:70 | |
| MEDIUM | Potential Arbitrary File Read via `documentPath` The `create_project` A2A endpoint supports a `fromFile` mode with a `documentPath` parameter, allowing the skill to read a local file to generate project data. If the TensorPM application reads arbitrary local files specified by this path without proper validation or sandboxing, there's a potential risk of data exfiltration or exposure of sensitive local files if a malicious agent or local process can control this `documentPath`. The description does not specify any path validation or sandboxing mechanisms. Implement strict path validation and sandboxing for `documentPath` to ensure only intended files (e.g., within a specific user-approved directory or a temporary upload location) can be accessed. Ensure that the content of such files is not inadvertently exposed or exfiltrated through the API or AI responses. | LLM | SKILL.md:190 |
Scan History
Embed Code
[](https://skillshield.io/report/60264302ab71d5d9)
Powered by SkillShield