Security Audit

here-be-git

github.com/intellectronica/agent-skills

AI SkillCommit 9b0e00ad1b94

CAUTION

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

here-be-git received a trust score of 72/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.

SkillShield's automated analysis identified 1 finding: 1 critical, 0 high, 0 medium, and 0 low severity. Key findings include Persistent Agent Behavior Manipulation via AGENTS.md.

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 June 1, 2026 (commit 9b0e00ad). 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

70%

Behavioral Risk Signals

Filesystem Write

1 finding

Dynamic Code

1 finding

Security Findings1

Severity	Finding	Layer	Location
CRITICAL	Persistent Agent Behavior Manipulation via AGENTS.md The skill creates or modifies an `AGENTS.md` file containing highly imperative and manipulative instructions intended to dictate the agent's future behavior, specifically regarding git commit practices. This content uses strong, absolute language ("CRITICAL, MANDATORY, NO EXCEPTIONS", "THIS IS NOT OPTIONAL.", "MUST follow", "ZERO exceptions", "ALWAYS commit") to enforce a specific workflow. This constitutes a persistent prompt injection attempt, as the agent is likely to encounter and be influenced by this file in subsequent operations within the repository, potentially overriding its discretion or direct user instructions. Avoid creating files with content that attempts to dictate or override the agent's core operational instructions or decision-making process. If guidance is needed, it should be framed as suggestions or best practices, not mandatory, exception-less rules. The agent's primary directive should always come from the current user interaction, not from persistent files created by previous skill executions. Such strong directives can lead to an agent ignoring user intent or becoming overly rigid.	LLM	SKILL.md:22

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/9e40275df8bdd21e.svg)](https://skillshield.io/report/9e40275df8bdd21e)