The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests video transcripts and raw recordings to perform script compression and narration regeneration, creating a surface for indirect prompt injection where instructions could be embedded in the transcript data.\n
Ingestion points: Processes existing Clueso projects or uploaded raw screen recordings (transcript reading and runtime estimation step in SKILL.md).\n
Boundary markers: No specific boundary markers or instructions to ignore embedded commands in the transcript are present.\n
Capability inventory: Capability to write to Clueso projects, regenerate audio narration, re-align visuals, and generate review links.\n
Sanitization: No evidence of sanitization or filtering of the transcript content before it is used to iterate on script versions.\n- [EXTERNAL_DOWNLOADS]: The skill provides instructions to connect the agent to an external Model Context Protocol (MCP) server for functionality.\n
Evidence: Instructs the connection to https://connect.clueso.io/mcp as a vendor-provided resource for video editing capabilities.\n- [COMMAND_EXECUTION]: The skill instructs the agent to provide specific shell commands to the user to configure the execution environment.\n
Evidence: Instructs suggesting the command claude mcp add --transport http Clueso https://connect.clueso.io/mcp to the user in Claude Code environments.

tighten-narration