The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to perform installation, configuration, and operational tasks, including binary execution for the instacart-pp-cli tool.
[REMOTE_CODE_EXECUTION]: The instructions direct the agent to install external software using npx -y @mvanhorn/printing-press-library and go install github.com/mvanhorn/printing-press-library/.... These commands download and execute code from the author's infrastructure.
[EXTERNAL_DOWNLOADS]: The skill's 'Backfill Flow' uses WebFetch to download multiple JavaScript files from https://raw.githubusercontent.com/mvanhorn/printing-press-library/... which are then intended to be injected into a live browser session using a JavaScript injection tool.
[CREDENTIALS_UNSAFE]: The skill manages sensitive Instacart session cookies, extracting them from Chrome and storing them locally in a configuration file at ~/.config/instacart/session.json. While documented, this file contains active authentication tokens.
[INDIRECT_PROMPT_INJECTION]: The skill processes order history data from Instacart, creating a vulnerability surface where instructions embedded in purchase history could influence agent behavior.
Ingestion points: The instacart history import command processes JSONL files containing external order data into a local database.
Boundary markers: No explicit boundary markers or 'ignore' instructions are specified for the imported data processing.
Capability inventory: The skill possesses the ability to execute shell commands, fetch remote files, and inject scripts into browser contexts.
Sanitization: The instructions do not define specific sanitization or validation routines for the external order data before it is indexed for item resolution.

pp-instacart