skills/aradotso/security-skills/sparkfinderoven-claude-security-compliance-suite/Gen Agent Trust Hub
sparkfinderoven-claude-security-compliance-suite
Warn
Audited by Gen Agent Trust Hub on May 29, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the user to clone the project repository from an external source (
https://github.com/sparkfinderoven/r01-hesreallyhim-awesome-claude-code-security.git) during installation. - [COMMAND_EXECUTION]: The incident response playbooks within the skill include high-impact system commands such as
iptablesfor network isolation,ddfor forensic disk imaging, and AWS CLI operations (aws ec2 modify-instance-attribute) for infrastructure modification. - [DATA_EXFILTRATION]: The skill requests the configuration of sensitive environment variables, including
GITHUB_TOKEN,AWS_PROFILE,SMTP_PASSWORD, andNVD_API_KEY. It also defines integrations for outbound communication via Slack webhooks and SMTP, which could be exploited for data exfiltration if the agent is compromised. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its design:
- Ingestion points: Processes untrusted codebase files, dependency manifests, architecture diagrams, and security logs as part of its scanning and auditing functions.
- Boundary markers: The instructions lack explicit delimiters or warnings to prevent the agent from following instructions embedded within the data being scanned.
- Capability inventory: The skill possesses extensive capabilities, including file system read/write access, shell command execution (including privileged system commands), and network access.
- Sanitization: No sanitization or validation logic is provided to filter malicious instructions from the untrusted data before it is analyzed by the agent.
Audit Metadata