The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes transcript data from external phone calls, creating a surface for indirect prompt injection.
Ingestion points: Call transcripts are retrieved using get_call_result (MCP) or callsimple result (CLI) in SKILL.md.
Boundary markers: The agent is instructed to report findings in a few lines, but there are no specific markers to isolate the transcript content from instructions.
Capability inventory: The skill allows placing calls, fetching results, and executing shell commands via the CLI.
Sanitization: There is no evidence of sanitization for the transcript data before it is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill fetches the callsimple-agent CLI tool from the author's GitHub repository (github:IndianAppGuy/callsimple-agent) using npx or npm.
[REMOTE_CODE_EXECUTION]: The skill uses npx to execute code directly from a remote GitHub repository for authentication and call management.
[COMMAND_EXECUTION]: The skill utilizes shell commands (npx, npm, callsimple) to interact with the phone call service and manage account balances.

callsimple