# Simplify Scope

> A Claude skill from 0-uddeshya-0/agent-skills.

- **Type:** Skill
- **Install:** `agentstack add skill-0-uddeshya-0-agent-skills-simplify-scope`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [0-uddeshya-0](https://agentstack.voostack.com/s/0-uddeshya-0)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [0-uddeshya-0](https://github.com/0-uddeshya-0)
- **Source:** https://github.com/0-uddeshya-0/agent-skills/tree/main/skills/simplify-scope

## Install

```sh
agentstack add skill-0-uddeshya-0-agent-skills-simplify-scope
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# skill: simplify-scope

## purpose
Cut a bloated task to the minimum version that still satisfies the user's actual goal. Prevents agents from doing 8 things when 2 were needed.

## when_to_use
- Task has accumulated scope through a long conversation ("also do X, and while you're at it...")
- Request contains nested sub-tasks that weren't explicitly asked for
- Agent output will clearly be excessive for the actual request

## when_not_to_use
- Task is already atomic and specific
- User explicitly asked for exhaustive or comprehensive output
- Cutting scope would change the core answer

## input
```
task: string          # Full task as currently understood
user_goal: string     # What success looks like — the actual objective
```

## output
```json
{
  "simplified_task": "string",
  "removed": ["item — reason"],
  "kept": ["item — reason"],
  "reduction": "none" | "minor" | "significant"
}
```

## instructions

```
If user_goal is empty: infer it from the task itself — identify the single outcome the user most likely needs and treat that as the goal.

Cut scope, not quality. Find the minimum task that makes the user satisfied.

For each part of the task ask:
- Did the user explicitly request this, or was it assumed?
- Can the user do this in 30 seconds after seeing the core output?
- Does this require a separate action that wasn't requested?

If any answer is yes: remove it.
Never remove anything the user explicitly stated.
One-line justification for each removal and each kept item.

Return JSON only.
```

## constraints
- Max output tokens: 250
- simplified_task must be completable in a single response
- Do not remove explicitly stated requirements

## example

**Input:**
```
task: "Write a blog post about Python async with intro to concurrency, history of threading, GIL explanation, Node.js comparison, asyncio examples, benchmarks, pitfalls, migration guide, and FAQ."
user_goal: "Help our team understand when to use async."
```

**Output:**
```json
{
  "simplified_task": "Write a practical post explaining Python async/await: what it is, when to use it vs threads, and 2 code examples. For engineers already comfortable with Python.",
  "removed": [
    "Threading history — background, not needed to understand async",
    "GIL deep-dive — tangential to the goal",
    "Node.js comparison — adds scope without serving stated goal",
    "Benchmarks — requires running code, separate deliverable",
    "Migration guide — separate article",
    "FAQ — add based on reader questions later"
  ],
  "kept": [
    "Async/await explanation — core to the goal",
    "When to use vs threads — directly answers the team need",
    "Code examples — essential for practical understanding"
  ],
  "reduction": "significant"
}
```

## feedback_log

# 2026-05-29 | invoked with user_goal left blank | skill had no fallback — stalled on missing required field | added: infer goal from the task itself when user_goal is not provided

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [0-uddeshya-0](https://github.com/0-uddeshya-0)
- **Source:** [0-uddeshya-0/agent-skills](https://github.com/0-uddeshya-0/agent-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-0-uddeshya-0-agent-skills-simplify-scope
- Seller: https://agentstack.voostack.com/s/0-uddeshya-0
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.