Wrapper around cursor-agent CLI for non-interactive runs (prompt via args/file/stdin) with resume support.
From agentic-coding-toolsnpx claudepluginhub thesylvester/agentic-coding-tools --plugin read-transcriptThis skill is limited to using the following tools:
scripts/cursor-agentSearches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Thin wrapper around cursor-agent CLI:
PROMPT_FILE, or stdin--resume or CURSOR_SESSION env var--force, --approve-mcps, --browser# Arguments
.claude/skills/cursor-agent/scripts/cursor-agent Your prompt here
# File via env var
PROMPT_FILE=task.md .claude/skills/cursor-agent/scripts/cursor-agent
# Stdin
cat task.md | .claude/skills/cursor-agent/scripts/cursor-agent
# Resume by session ID (returned from previous run)
.claude/skills/cursor-agent/scripts/cursor-agent --resume <session-id> "Follow-up question"
# Via environment variable
CURSOR_SESSION=<session-id> .claude/skills/cursor-agent/scripts/cursor-agent "Follow-up"
# Via environment variable
CURSOR_AGENT_MODEL=gpt-5 .claude/skills/cursor-agent/scripts/cursor-agent "Your prompt"
| Variable | Description |
|---|---|
PROMPT_FILE | Read prompt from this file |
CURSOR_AGENT_MODEL | Passed as --model |
CURSOR_SESSION | Session ID to resume (fallback: CHAT_ID, SESSION_ID) |
Returns the response followed by the session ID:
<response text>
[session_id: <uuid>]
Capture the session ID to continue the conversation in subsequent calls.
cursor-agent --print --output-format json in non-interactive mode--fork/-f and CHAT_ID/SESSION_ID still work