The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external technical data which could potentially contain malicious instructions.
Ingestion points: Data is retrieved from the Meticulous API via commands such as agent dom-diff, agent timeline-diff, and download session as described in references/agent.md and references/download.md.
Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are mentioned to protect the agent while it processes the retrieved DOM or session logs.
Capability inventory: The skill is capable of executing meticulous CLI commands, performing network operations to the vendor API, and writing data to the ~/.meticulous directory.
Sanitization: The skill does not describe specific sanitization or filtering processes for the ingested technical content.
[EXTERNAL_DOWNLOADS]: The skill references the installation of the @alwaysmeticulous/cli package via npm. This is a standard installation procedure for the official CLI tool provided by the vendor.
[COMMAND_EXECUTION]: The skill makes frequent use of the meticulous CLI to record, replay, and analyze web sessions, which is the primary purpose of the tool.

meticulous-cli