Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/ningzio/cursor-plugin-cc/cursor-dispatchgit clone --depth 1 https://github.com/ningzio/cursor-plugin-ccWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00025 | $0.00973 |
| Opus 5 | $0.00013 | $0.00487 |
| Sonnet 5 | $0.00005 | $0.00195 |
| Haiku 4.5 | $0.00003 | $0.00097 |
Grade A, and why
cursor-dispatch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a thin forwarding wrapper around cursor-companion.mjs dispatch. Read this in full before invoking — the sandbox model is the part most callers miss.
Note for upstream callers (Claude Code main thread, Codex, other agents)
Before you delegate to this subagent, packing the prompt is YOUR job — this subagent only forwards.
Cursor's auto model is billed per request, not per token. One fat dispatch with 4M tokens costs the same as one thin dispatch with 40K tokens — both are 1 request. To maximize each call, bundle into one dispatch:
- Scope — files/dirs in-bounds and out-of-bounds.
- Acceptance criteria — tests to pass, invariants to hold, behaviors to verify.
- Regression checks — adjacent things to spot-check.
- Self-validation — ask cursor to run build/tests/lint and report results.
Avoid splitting one feature into 3–4 thin dispatches; bundle into one with a checklist. This applies only when --model auto is in effect (default). For explicit-model dispatches, billing is token-based — keep prompts lean.
See skills/cursor-cli-runtime/SKILL.md for the full guidance.
Sandbox model (READ THIS — v0.3+ changed the default)
- Default = in-place. Cursor edits the caller's cwd directly. Commits land on the caller's current branch. No cherry-pick needed, no
.cursor/worktrees/. --isolatedopts into the legacy sandbox: cursor runs in.cursor/worktrees/<jobId>/on a<jobId>branch and auto-commits on success. Commits do NOT appear on the caller's branch — caller must cherry-pick if they want them.--backgrounddefaults to--isolated(a background agent + live edits in cwd = corruption). Pass--in-placeto override.- A prompt like "stay on branch X" has no effect. The branch decision is made by this wrapper based on
--isolated/--in-placeflags, not by cursor-agent. Set the flags correctly upstream. - If dispatch refuses with a
REFUSED: <CODE>line on stderr, that's a policy refusal with concrete next-step suggestions. Read theCaller next steps:list and act accordingly — do not retry blindly with the same flags.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 52 lines · 25 tokens per session scan A edfd1540e36c
cursor-dispatch is an agent published in the GitHub repository ningzio/cursor-plugin-cc (2 stars, last pushed 3mo ago), licensed MIT. It adds 25 tokens to every session and 973 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.