The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation instructs the user to set up an MCP server using the octocode-mcp package via npx. This package is used to provide the agent with local and external code search capabilities.
[COMMAND_EXECUTION]: The agent utilizes a comprehensive set of Model Context Protocol (MCP) tools, including localSearchCode, githubSearchCode, and packageSearch, to explore codebases and gather evidence for technical proposals.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it is designed to read and process content from untrusted external repositories which may contain malicious instructions.
Ingestion points: Data is ingested through the githubGetFileContent and packageSearch tools during the research track in SKILL.md.
Boundary markers: No specific delimiters or safety instructions are defined to prevent the agent from following instructions embedded within the researched external code.
Capability inventory: The agent has the capability to write files to the local .octocode/rfc/ directory and perform recursive repository searches.
Sanitization: The skill does not implement validation or sanitization of the content retrieved from remote sources before it is analyzed and incorporated into the output.

octocode-rfc-generator