The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The documentation link https://jeffallan.github.io/claude-skills/skills/backend/dotnet-core-expert/ and its root domain are flagged as malicious by automated scanners. Providing documentation via a blacklisted domain is a significant security risk.
[METADATA_POISONING]: The skill's metadata contains an author link to a GitHub profile and a documentation site associated with a blacklisted infrastructure. Additionally, the skill file SKILL.md itself has been flagged by file reputation scanners (FileRepMalware), indicating a high probability of association with malicious activity.
[COMMAND_EXECUTION]: The workflow in SKILL.md directs the agent to execute dotnet build, dotnet test, and curl commands. While these are standard tools for .NET development, they provide a vector for arbitrary code execution if the skill processes malicious code or requirements.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user requirements and data models to generate architectural patterns and code, creating an attack surface for indirect prompt injection.
Ingestion points: SKILL.md core workflow step 1 (Analyze requirements, identify architecture pattern, data models).
Boundary markers: Absent; there are no instructions to ignore or delimit potentially malicious directives embedded in the input requirements.
Capability inventory: The skill utilizes dotnet build, dotnet test, and curl as part of its operational workflow.
Sanitization: Absent; there is no mention of validation, filtering, or escaping for external content before it is processed for code generation.

dotnet-core-expert