The Agent Skills Directory

[PROMPT_INJECTION] (HIGH): The skill is vulnerable to Indirect Prompt Injection. It processes untrusted data from user-controlled Markdown files and possesses the capability to modify the filesystem when the --fix argument is used. Evidence Chain: 1. Ingestion points: Reads .md and .qmd files from the workspace as defined in the usage examples. 2. Boundary markers: Absent; there are no instructions provided to the agent to treat file contents as untrusted data or to ignore embedded instructions. 3. Capability inventory: Executes the markdownlint-cli2 command and can perform file-system writes via the --fix flag. 4. Sanitization: None; external file content is passed directly to the tool's logic.
[COMMAND_EXECUTION] (LOW): The skill relies on executing shell commands to perform its core functions. While this is the intended behavior for a linter, it serves as the mechanism for potential side effects if the agent is manipulated by malicious content within the files being processed.
[EXTERNAL_DOWNLOADS] (LOW): Recommends the installation of markdownlint-cli2 via npm or brew. Per the [TRUST-SCOPE-RULE], these are considered trusted package managers/registries, resulting in a Low severity for the download recommendation itself.

markdownlint