The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the user to execute standard administrative and development tools, such as ps, kill, docker, and wrangler, to manage and inspect running processes during diagnostic procedures.
[EXTERNAL_DOWNLOADS]: The documentation suggests installing the websockets Python library. This is a well-known, legitimate package required for the skill's core functionality of establishing communication with the Node.js inspector.
[DYNAMIC_EXECUTION]: Memory analysis is performed by executing JavaScript snippets within the target runtime using the Chrome DevTools Protocol (Runtime.evaluate). This is an intentional and necessary feature for live process inspection and heap analysis.
[INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion surface through the processing of external telemetry data.
Ingestion points: OpenTelemetry logs are queried from a ClickHouse database as described in SKILL.md.
Boundary markers: No specific delimiters are identified to separate log content from potential instructions.
Capability inventory: The skill environment permits shell command execution, WebSocket networking, and dynamic JavaScript evaluation across all provided scripts.
Sanitization: There is no explicit sanitization or validation of the ingested log data before it is presented to the agent or used in diagnostic workflows.

deco-site-memory-debugging