The Agent Skills Directory

[PERSISTENCE]: The skill provides instructions for the agent to modify the user's ~/.bashrc file by appending an export command. This ensures the Node.js binary path remains in the PATH environment variable across future shell sessions, which is necessary for the manually installed runtime to function correctly.
[EXTERNAL_DOWNLOADS]: The skill automates the retrieval of the Node.js binary and its SHA256 checksums from nodejs.org to facilitate environment bootstrapping when the runtime is missing. It also makes use of registry.npmmirror.com as a fallback registry for installing the required npm packages. Both sources are recognized as well-known and established services.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, which could lead to indirect prompt injection if the inputs contain adversarial instructions.
Ingestion points: User-provided image file paths for '模特图' (model) and '替换衣物图' (clothing), and the '提示词' (prompt) field in SKILL.md.
Boundary markers: None identified in the prompt interpolation logic to distinguish user data from system instructions.
Capability inventory: The skill executes the qhkit CLI tool, which interacts with the local file system and performs network operations to the vendor's API.
Sanitization: The skill utilizes a JSON parameter file (params.json) to pass data to the CLI tool, which effectively mitigates shell command injection but does not prevent logic-based injection within the AI workflow.
[COMMAND_EXECUTION]: The skill's primary functionality is driven by executing the qhkit CLI tool to perform various operations, including retrieving workflow options, calculating credit costs, and generating the final image output.

qinghu-model-outfit-restore