The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs shell-based operations to manage project dependencies and verify implementation quality.
Executes platform-specific package manager commands such as npm install, pip install, pnpm start, and bundle install to integrate the PostHog SDK (Step 3).
Invokes project-defined build scripts, type checkers, and formatting tools like Prettier during the verification phase (Step 10).
[EXTERNAL_DOWNLOADS]: Fetches software components and configuration patterns from established technology providers.
Downloads SDK packages from official public registries (NPM, PyPI, RubyGems).
References implementation examples and technical documentation hosted on PostHog's official GitHub repositories and web domains.
[DATA_EXFILTRATION]: Configures the application to transmit telemetry data and user identifiers to an external analytics platform.
Instruments the codebase to capture user behavioral events and identification metadata, such as usernames or email addresses extracted from login and signup forms (Step 5, Step 6, Step 7).
Transmits the collected data to PostHog's ingestion infrastructure. This behavior is the intended function of the skill for providing product analytics.
[SAFE]: The skill analyzes codebase files and writes configuration, presenting a surface for indirect prompt injection that is inherent to its instrumentation purpose.
Ingestion points: Reads local project structure, dependency files (e.g., package.json, requirements.txt), and source code to plan instrumentation (Step 1, Step 5).
Boundary markers: Absent.
Capability inventory: Performs file system writes and executes shell commands for package installation and linting (Step 3, 6, 7, 8, 9, 10).
Sanitization: Absent.

instrument-product-analytics