The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the user to install the core CLI tool using a high-risk command that pipes a remote script directly into the bash shell.\n
Evidence: curl -fsSL https://raw.githubusercontent.com/danielmiessler/fabric/main/scripts/installer/install.sh | bash found in SKILL.md.\n- [COMMAND_EXECUTION]: The skill relies on the execution of multiple system commands and third-party tools to perform its functions, which are carried out via the Bash tool.\n
Evidence: Commands such as fabric, brew, winget, git, and npm are used throughout SKILL.md and SKILL.toon.\n- [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by design, as it pipes potentially malicious or untrusted external data directly into Large Language Model prompts.\n
Ingestion points: Data is ingested from external URLs (via curl), system logs (/var/log/app.log), and development tool outputs (git diff, npm test) as seen in SKILL.md steps 2 and 7.\n
Boundary markers: The provided usage examples do not include delimiters or system instructions to ignore embedded commands within the piped input.\n
Capability inventory: The skill has permissions to execute shell commands (Bash tool) and perform network requests.\n
Sanitization: No evidence of input sanitization, escaping, or validation is present in the skill's instructions.

fabric