The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill analyzes project files such as package.json and Cargo.toml to detect frameworks and configurations, creating a surface for indirect prompt injection. 1. Ingestion points: Project metadata and source code files identified in Phase 1 and 2 detection steps. 2. Boundary markers: Absent; no specific instructions to treat project content as untrusted. 3. Capability inventory: The skill performs file writing to instrument applications and executes command-line package installations. 4. Sanitization: Absent; there are no instructions to validate or filter project-extracted strings before interpolation.
[EXTERNAL_DOWNLOADS]: The skill references official vendor packages including @outlit/browser, @outlit/node, and the outlit Rust crate. It also includes instructions for self-installation from the OutlitAI/skills GitHub repository via npx.
[COMMAND_EXECUTION]: The skill uses package manager commands (such as bun add or npm install) and CLI tools to install dependencies and configure the integration environment based on user project state.

outlit-sdk