The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the gh CLI to interact with GitHub Projects and uses shell commands for cache management and git branch detection. Additionally, it automates the installation of a persistent shell hook into .claude/settings.local.json. This hook executes complex shell logic (git, grep, sed, date) every time a user submits a prompt in that environment.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by processing and displaying data from GitHub Projects.
Ingestion points: Fetches item titles and fields from GitHub via the gh CLI and stores them in local JSON files (.ghpm/cache.json, .ghpm/sessions/*.json).
Boundary markers: The output formatting conventions in references/format.md do not include delimiters or instructions for the agent to ignore embedded commands in the external data.
Capability inventory: The skill executes shell commands and modifies agent-level configuration files.
Sanitization: There is no evidence of sanitization or escaping of project content before it is presented to the AI agent or processed by the integration hooks.

ghpm-shared