PM Skills is a collection of plain-Markdown instructions that teach AI assistants structured methods for handling professional, personal, and life-admin tasks. People use it with Claude, ChatGPT, Gemini, Cursor, Codex, and other supported agents for work such as writing product requirements, reviewing documents, or planning difficult situations.
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.
git clone --depth 1 https://github.com/mohitagw15856/pm-claude-skillsWrote 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/rules/mohitagw15856/pm-claude-skills/delegate-to-ai)<a href="https://agentmods.dev/rules/mohitagw15856/pm-claude-skills/delegate-to-ai"><img src="https://agentmods.dev/badge/rules/mohitagw15856/pm-claude-skills/delegate-to-ai/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/mohitagw15856/pm-claude-skills/delegate-to-ai"><img src="https://agentmods.dev/badge/rules/mohitagw15856/pm-claude-skills/delegate-to-ai.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00133 | $0.01129 |
| Opus 5 | $0.00067 | $0.00564 |
| Sonnet 5 | $0.00027 | $0.00226 |
| Haiku 4.5 | $0.00013 | $0.00113 |
Grade A, and why
delegate-to-ai 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 7d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegate to AI
The useful frame for AI is a new hire: fast, tireless, capable of a lot — and unreliable in specific ways, so you delegate deliberately and verify the important stuff. The mistake is either delegating nothing (no leverage) or delegating things that need your judgment (lost quality, lost edge). This sorts your actual workload into what to fully hand over, what to hand over with review, and what to keep — with the reasoning, so you free your time on purpose.
What This Skill Produces
- A three-way sort of your tasks — 🟢 delegate fully (low-stakes, AI does it well) · 🟡 delegate with review (AI drafts, you approve) · 🔴 keep human (judgment, relationships, high-stakes)
- The reasoning per line — why each task landed where it did, so the map makes sense and you can adjust it
- How to brief the delegated ones — what the AI needs to do the handed-off tasks well (context, examples, the output you want)
- Where your judgment is the value — the tasks where you are the point, that delegating would hollow out
- A time-back estimate — a rough sense of what delegating frees up, so it's worth doing
- A start-here pick — the one or two tasks to delegate first to build the habit and trust
Required Inputs
Ask for these if not provided:
- Your tasks — the recurring things on your plate (a brain-dump is fine)
- Your role & what you're good at — so we protect the judgment work that's your edge
- The stakes per task — what errors cost (drives review level)
- Your AI access — what tools you have, so the delegation is realistic
- What's draining you — the time-sinks you'd most love off your plate
Framework: Manage AI Like A Capable New Hire
- List the real workload. Get the tasks out of your head — you can't delegate what you haven't named.
- Rate each on two axes. How well AI does it × how much errors cost. High-ability/low-stakes → delegate fully. High-ability/high-stakes → delegate with review. Needs-human-judgment → keep.
- Protect your edge. Explicitly flag the tasks where your judgment, relationships, or taste are the value — delegating those doesn't save time, it removes the point.
- Brief the delegated tasks. For the ones you hand over, define what the AI needs — context, examples, output format — like onboarding a new hire, not tossing over a wall.
- Set the review level. Full-delegate = spot-check. Delegate-with-review = you approve every output before it's used, especially anything external or irreversible.
- Start small, build trust. Pick one or two to delegate first; verify the quality; expand as trust builds.
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.
- 7d ago First seen · 69 lines · 133 tokens per session scan A 94992ae493dc
delegate-to-ai is a cursor rule published in the GitHub repository mohitagw15856/pm-claude-skills (1,352 stars, last pushed 3d ago), licensed MIT. It adds 133 tokens to every session and 1,129 once invoked, about $0.0007 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 2026-09-03.
Other cursor rules, from other repositories
cursor
Always-on my-adhd-caveman. Off with "stop adhd-caveman" or "normal mode".
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.