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 skills/zebbern/agent-collab/cursor-delegationnpx skills add zebbern/agent-collab --skill cursor-delegationgit clone --depth 1 https://github.com/zebbern/agent-collabWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zebbern/agent-collab/cursor-delegation)<a href="https://agentmods.dev/skills/zebbern/agent-collab/cursor-delegation"><img src="https://agentmods.dev/badge/skills/zebbern/agent-collab/cursor-delegation.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00066 | $0.01120 |
| Opus 5 | $0.00033 | $0.00560 |
| Sonnet 5 | $0.00013 | $0.00224 |
| Haiku 4.5 | $0.00007 | $0.00112 |
Grade A, and why
cursor-delegation 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 5d ago.
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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cursor Delegation
This skill is the policy and the loop for delegating work to Cursor without a /cursor:* command. Presenting output stays governed by the cursor-result-handling skill; prompt shape and model notes stay governed by the cursor-prompting skill. Do not duplicate either — defer to them.
When to delegate — and when not
- Delegate when parallel speed adds real value: an implementation or scaffolding job that can run while the session continues on something else, or an everyday review of a substantive diff.
- Evidence note (portfolio retro, 2026-08-07): across the first 8 ledger dispositions Cursor was never exercised as a delegate, so its routing guidance rests on less evidence than Codex's — when an increment genuinely fits the profile above, prefer delegating it here over defaulting to local work, so the ledger accumulates the comparison the routing policy needs.
- Keep trivial work local. A rename, a one-file edit, a question Claude can answer directly — spawning a background agent there is overhead, not help.
- Non-transferable context, 2026-08-08 (docs/bench-pilot-2026-08-08.md): a bench pilot measured the CODEX arm matching solo Claude's fix rate on one well-localized single-file bugfix cell while costing ~30% more wall time. No Cursor arm has been benched, so that result transfers no verdict here — the evidence note above stands, and bugfix-shaped increments are among the shapes a future Cursor arm should measure.
- Never delegate when the user explicitly asked Claude to do the work personally.
- Run one delegated job of a class at a time (one task, one review). Before firing a new one, run
status --jsonand confirm nothing of the same class is still active.
Disclosure
- Announce every delegation in one short line when starting it — what is being delegated and that it is going to Cursor. Never silently spawn CLI work.
The loop: fire, await, collect
Fire — start the job in the background:
node "${CLAUDE_PLUGIN_ROOT}/scripts/cursor-companion.mjs" task --background [--write] [--profile deep|fast] [--model <model>] "<prompt>"
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.
- 5d ago First seen · 78 lines · 66 tokens per session scan A d8348e0017ad
cursor-delegation is a skill published in the GitHub repository zebbern/agent-collab (32 stars, last pushed 20d ago), licensed Apache-2.0. It adds 66 tokens to every session and 1,120 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
team-lead-reference
Provides model routing rules, validates delegation prerequisites, supplies cost tracking templates, defines dead-letter queue formats for Team Lead orchestration. Load when assigning tasks to agents, choosing model tiers, starting delegation session, running multi-agent workflow, delegating work, choosing which model…
vitest-testing
Vitest unit and integration testing patterns, commands, mocking (vi.mock), and coverage. Use when writing .test.ts files, configuring the test runner, or adding coverage thresholds.
orchestration-protocols
Coordinate multiple agents: parallel spawning, health monitoring, circuit breakers, escalation. Use for parallel agents, agent timeouts, fan-out tasks, multi-agent delegation.
typescript-best-practices
Type-level discipline for TypeScript: discriminated unions, branded types, constructive modeling, narrowing hierarchy, exhaustiveness checks, boundary validation. Use when reading or editing any .ts or .tsx file, modelling domain types, or refactoring an as cast out of existing code.
turborepo-monorepo
Configure pipelines, set up local/remote caching, and run scoped tasks in a Turborepo monorepo. Use when you say: 'enable remote caching', 'optimize pipeline inputs/outputs', or 'filter builds to affected packages'.
agent-hooks
Four lifecycle hooks every agent runs: on-session-start scans LESSONS-LEARNED.md, resumes checkpoints; on-pre-delegate verifies tracker issues, file partitions; on-post-delegate runs fast-review, CI checks; on-session-end runs the health checks, writes logs. Use when starting a new session, running pre-flight checks…