Security Audit

harness/harness-skills:skills/configure-agent-pr-attestation

github.com/harness/harness-skills

AI SkillCommit 770a4427aa4a

TRUSTED

Scanned about 1 month ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

harness/harness-skills:skills/configure-agent-pr-attestation received a trust score of 79/100, 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 Sensitive environment variable access: $HOME, Command Injection via Unsanitized Template Substitution in Hook Script.

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 10, 2026 (commit 770a4427). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

93%

Dependency Graph

100%

LLM Behavioral Safety

85%

Behavioral Risk Signals

Network Access

1 finding

Shell Execution

2 findings

Dynamic Code

1 finding

Excessive Permissions

1 finding

Security Findings2

Severity	Finding	Layer	Location
HIGH	Command Injection via Unsanitized Template Substitution in Hook Script The skill generates a shell script (`agent-pr-attestation.sh`) by directly substituting `{{COMMAND_FILTER}}` and `{{URL_REGEX}}` into single-quoted bash variables. If an attacker or a malicious repository configuration manipulates the agent into using a custom command filter or URL regex containing single quotes (e.g., `foo'; malicious_command; '`), it will break out of the shell quotes and execute arbitrary commands every time a tool is used (via the `PostToolUse` hook). Sanitize or escape single quotes in `COMMAND_FILTER` and `URL_REGEX` before substituting them into the shell script, or pass these values via environment variables or a structured configuration file (e.g., JSON) rather than generating dynamic shell code.	LLM	SKILL.md:115
MEDIUM	Sensitive environment variable access: $HOME Access to sensitive environment variable '$HOME' detected in shell context. Verify this environment variable access is necessary and the value is not exfiltrated.	Static	skills/configure-agent-pr-attestation/SKILL.md:104

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/55b0317704b8fb2d.svg)](https://skillshield.io/report/55b0317704b8fb2d)