The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The documentation URL https://jeffallan.github.io/claude-skills/skills/backend/dotnet-core-expert/ included in SKILL.md is flagged as malicious by automated scanners (URLite Blacklist). \n- [METADATA_POISONING]: The file SKILL.md has been flagged by file reputation scanners as potentially malicious (FileRepMalware). \n- [METADATA_POISONING]: The skill metadata lists the author as Jeffallan, which contradicts the platform-provided author identity of sodiqabdulwaris. This discrepancy is a strong indicator of deceptive packaging or impersonation. \n- [COMMAND_EXECUTION]: The core workflow requires the execution of shell commands such as dotnet build and dotnet test. While standard in the .NET ecosystem, these commands pose a high risk when executed in an environment associated with malicious file reputations and blacklisted URLs. \n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external requirements and code files without sufficient isolation or validation, making it vulnerable to indirect injection. \n
Ingestion points: Project requirements, API designs, and C# source code provided by the user and processed by the agent. \n
Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore instructions embedded in the user project. \n
Capability inventory: Extensive use of shell commands through the dotnet CLI, which allows for subprocess execution. \n
Sanitization: Absent; the skill lacks any mechanism to validate or escape external input before processing it via instructions or commands.

dotnet-core-expert