The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of an external CLI tool via the command npm install -g @aicontextlab/cli or via npx. This package is hosted on the public NPM registry and is not part of the provided trusted vendors list.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute various CLI commands (oc init, oc folder, oc doc, oc search, etc.) and to modify configuration files in IDE-specific directories such as ~/.cursor/mcp.json and ~/.claude/commands.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by retrieving and processing external data from the file system to serve as context for the agent.
Ingestion points: Data is ingested from document files stored in ~/.opencontext/contexts and from the local repository via tools like oc_search, oc_list_docs, and oc_manifest.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands were found in the skill's provided workflows.
Capability inventory: The agent has access to Bash, Write, Read, Grep, and Glob tools, which could be leveraged if it follows malicious instructions embedded in the retrieved context.
Sanitization: There is no evidence of sanitization or content filtering for the data retrieved from the OpenContext store before it is presented to the LLM.

opencontext