The Agent Skills Directory

[SAFE]: The skill uses standard local development utilities including Grep, Glob, and the Dart/Flutter CLI tools for static code analysis. No evidence of unauthorized network access, credential harvesting, or remote code execution was found. All shell commands identified in the methodology are common for Dart project analysis (e.g., 'flutter pub deps', 'dart analyze').- [PROMPT_INJECTION]: The skill reads external project files into the agent context, creating a potential surface for indirect prompt injection. Ingestion points: Source code files in the 'lib/' and 'test/' directories, and the 'pubspec.yaml' configuration file. Boundary markers: The instructions do not provide explicit delimiters or instructions to ignore embedded commands within the analyzed code. Capability inventory: The skill utilizes 'Read', 'Glob', 'Grep', and 'Bash' tools to inspect the local filesystem. Sanitization: No specific sanitization or validation of the ingested code content is performed. This is an inherent risk of code-analysis skills and is not indicative of malicious intent.

dedup-code-agent