The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the simplified-apikit Python package from a custom PyPI registry hosted on Gitlab (https://gitlab.com/api/v4/projects/70495826/packages/pypi/simple). Gitlab is recognized as a well-known and reputable service provider, and the download is essential for the core functionality of the skill.
[COMMAND_EXECUTION]: The skill provides numerous examples of shell command execution using the smp CLI tool. These examples include piping data into python3 -c snippets to perform JSON parsing and variable extraction (e.g., import json,sys; print(json.load(sys.stdin)['result']['image_url'])). These are standard patterns for CLI integrations.
[DATA_EXFILTRATION]: The skill transmits user-provided media URLs and task payloads to the official Simplified API (api.simplified.com). This behavior is consistent with the skill's stated purpose of providing an interface to Simplified's cloud services.
[PROMPT_INJECTION]: The skill implements a 'Memory Pattern' that encourages the agent to cache data retrieved from the Simplified API—such as board IDs, status names, and user lists—within the agent's persistent memory. This creates a surface for indirect prompt injection if the data stored in the Simplified platform (e.g., a task title) contains malicious instructions designed to be executed when the agent later retrieves that data from its cache.
Ingestion points: JSON responses from the Simplified API retrieved via smp commands.
Boundary markers: None identified in the instructions for separating cached data from subsequent commands.
Capability inventory: Full access to the smp CLI and local shell execution.
Sanitization: None specified in the provided JSON parsing examples.

simplified-cli