Security Audit

PabloLION/bmad-plugin:plugins/bmad/skills/bmad-agent-sm

github.com/PabloLION/bmad-plugin

AI SkillCommit 17efb6ce67d3

TRUSTED

Scanned about 13 hours ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

PabloLION/bmad-plugin:plugins/bmad/skills/bmad-agent-sm received a trust score of 85/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.

SkillShield's automated analysis identified 1 finding: 0 critical, 0 high, 1 medium, and 0 low severity. Key findings include Broad filesystem search for project context.

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 April 11, 2026 (commit 17efb6ce). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

93%

Behavioral Risk Signals

Filesystem Write

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
MEDIUM	Broad filesystem search for project context The skill is instructed to search the entire filesystem (`/project-context.md`) for a file and load its content. This implies broad read access to the filesystem. If the `project-context.md` file contains sensitive information, or if the agent can be prompted to reveal the contents of other files found through this broad search mechanism, it poses a data exfiltration risk. The `/` wildcard allows searching in any subdirectory, which is a very permissive scope. 1. Restrict the search scope for `project-context.md` to a specific, known, and secure directory (e.g., `./project-context.md` or `docs/project-context.md`). 2. Ensure that the `project-context.md` file does not contain any sensitive information. 3. Implement strict output filtering or sanitization to prevent the agent from revealing raw file contents, especially from loaded context files, unless explicitly designed and secured for that purpose. 4. Consider if loading arbitrary files from the filesystem is truly necessary for the skill's function, or if a more controlled input mechanism (e.g., user upload, specific API call) could be used.	LLM	SKILL.md:45

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/25a8cc996cac8e58.svg)](https://skillshield.io/report/25a8cc996cac8e58)