security-scan
Warn
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to install the 'ecc-agentshield' package from the public npm registry. This package is maintained by an external developer (affaan-m) and is not from a recognized trusted organization or the skill's own author.
- [REMOTE_CODE_EXECUTION]: The use of 'npx ecc-agentshield' facilitates the immediate download and execution of remote code from the npm registry at runtime, which is a common vector for supply chain attacks.
- [COMMAND_EXECUTION]: The skill provides various commands to execute the external AgentShield binary. These commands run with the user's permissions and can read, analyze, and potentially modify ('--fix' flag) project files.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted content from local configuration files, creating a vulnerability to indirect prompt injection. Ingestion points: The tool specifically reads files in the '.claude/' directory, including 'CLAUDE.md' and 'mcp.json'. Boundary markers: The instructions do not specify any delimiters or safety markers to prevent the scanner from obeying instructions embedded within the files it audits. Capability inventory: The executed tool has read/write access to the local filesystem and network access for deep analysis features ('--opus' flag). Sanitization: There is no indication that the content from the configuration files is sanitized or validated before being processed.
Audit Metadata