The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions (in references/management.md) recommend installing the cocoindex-code-plus package from the public PyPI registry using uv, pipx, or pip. This is a vendor-provided tool required for the skill's functionality.
[COMMAND_EXECUTION]: The agent is instructed to execute the ccx CLI tool for various codebase exploration tasks, including search, grep, defs, refs, and ask. These commands interact with a remote query server over HTTP.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a bridge to external, potentially untrusted codebases. It fetches and processes code, documentation, and agent-generated summaries which could contain malicious instructions designed to subvert the agent's behavior.
Ingestion points: Content is ingested via subcommands like ccx search, ccx grep, ccx read-file, and the cited answers from ccx ask (referenced in SKILL.md).
Boundary markers: The skill documentation does not mention the use of specific delimiters or markers to distinguish between codebase content and agent instructions.
Capability inventory: The skill provides the agent with read-only access to remote codebase structures and file contents via the ccx CLI.
Sanitization: There is no evidence of sanitization, filtering, or escaping of the ingested codebase content before it is processed by the agent.

ccx-codebase-explorer