The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill includes a reference to an external documentation URL (https://jeffallan.github.io/claude-skills/skills/backend/rails-expert/) that has been identified as malicious (URL:Blacklist) by automated security scanners.
[REMOTE_CODE_EXECUTION]: The main skill instruction file SKILL.md has been flagged by file reputation scanners as containing potential malware (FileRepMalware).
[COMMAND_EXECUTION]: The skill facilitates the execution of various system commands, including rails generate, rails db:migrate, and bundle exec rspec, to implement Rails features and perform testing.
[METADATA_POISONING]: The skill metadata identifies the author as 'Jeffallan' and provides a corresponding GitHub link, which contradicts the platform-provided author information ('sodiqabdulwaris'), indicating potentially deceptive or misleading metadata.
[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes user-supplied requirements for code generation and command execution without sanitization.
Ingestion points: User requirements and specifications for Rails development tasks processed by instructions in SKILL.md.
Boundary markers: The skill does not provide delimiters or instructions for the agent to distinguish between its primary instructions and potentially malicious content within user requirements.
Capability inventory: The skill utilizes shell commands for rails generate, rails db:migrate, and bundle exec rspec across multiple files.
Sanitization: The instructions do not include steps to validate, escape, or sanitize user-provided inputs before they are incorporated into generated code or executed in the shell.

rails-expert