The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @continuedev/cli package from the NPM registry. This is the official command-line interface provided by the skill author and is a standard dependency for the described functionality.
[COMMAND_EXECUTION]: The skill uses shell commands to interact with the Continue CLI (cn check, cn login) and standard git operations (git apply). These commands are necessary for the primary purpose of auditing code changes and applying suggested fixes.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of local code changes (git diffs). This creates a surface for indirect prompt injection if an attacker embeds malicious instructions in code comments or strings. However, the risk is mitigated by the tool's design, which operates within temporary isolated worktrees and requires manual application of suggested patches.
Ingestion points: Git diff output from the local working tree, as described in the "How It Works" section of SKILL.md.
Boundary markers: Not explicitly defined in the general skill instructions; individual agent markdown files are responsible for their own prompt structure.
Capability inventory: The agent has permissions to read and modify files within a temporary git worktree to generate patches.
Sanitization: The skill relies on the agent's logic to differentiate between code and instructions; no specific input sanitization for the diff content is mentioned.

cn-check