The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reads and processes user-controlled Markdown notes (JDex entries) to automate filing, documentation, and information retrieval. This creates a surface where malicious instructions embedded in a note could influence the agent's behavior.
Ingestion points: Reads content from JDex entry notes (Markdown files) and follows links in metadata fields like "Related:" or "URL:".
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for the content within the processed files.
Capability inventory: Executes file system operations (ls, find), calls the "jd" CLI for moving files and creating IDs, and uses "bearcli" or "obsidian" for content search.
Sanitization: The instructions do not define any sanitization or validation of the note content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes several CLI tools to perform its functions, interacting directly with the host shell.
Evidence: Execution of the Johnny.Decimal CLI at "~/.local/share/johnnydecimal/cli/bin/jd", "bearcli", and the "obsidian" CLI. It also provides a command for adding an MCP server via "claude mcp add".
[EXTERNAL_DOWNLOADS]: The skill interacts with a remote MCP server and references a process for installing or updating local tools based on remote input.
Evidence: References the vendor's MCP server at "https://johnnydecimal.com/mcp" and describes following the "install_cli" process which involves following prompts from the server to update the local CLI.

jdex