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 skills add Cooper-X-Oak/goal-to-do --skill goal-todogit clone --depth 1 https://github.com/Cooper-X-Oak/goal-to-doWrote 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/cooper-x-oak/goal-to-do/goal-todo)<a href="https://agentmods.dev/skills/cooper-x-oak/goal-to-do/goal-todo"><img src="https://agentmods.dev/badge/skills/cooper-x-oak/goal-to-do/goal-todo/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/skills/cooper-x-oak/goal-to-do/goal-todo"><img src="https://agentmods.dev/badge/skills/cooper-x-oak/goal-to-do/goal-todo.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 | $0.00073 | $0.01444 |
| Opus 5.5 | $0.00029 | $0.00578 |
| Sonnet 5.5 | $0.00015 | $0.00289 |
| Haiku 4.5 | $0.00007 | $0.00144 |
Grade A, and why
goal-todo 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 2mo 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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Goal Todo
Overview
This skill turns a vague ambition into a final /goal prompt that is ready to paste and run.
Do not jump straight to writing the final prompt. First reduce ambiguity through structured interaction, then synthesize a durable long-running goal.
By default, the final user-facing deliverable should be the /goal prompt itself, not an explanatory brief.
Core Rule
When this skill is invoked, the default interaction surface should be:
- Use
request_user_inputin the current environment. - If
request_user_inputis unavailable, tell the user this skill works best with clickable structured choices and ask them to reopen or switch toPlan mode, then retry withrequest_user_input. - If the user does not switch to
Plan mode, fall back to concise plain-text alignment with explicit reply templates.
Do not skip the alignment round. Only the interaction surface may degrade.
Interaction Standard
Ask high-quality structured questions that reduce ambiguity instead of pushing the user into free-form ideation.
Match the user's current language by default. If the user has not established a clear language preference and the distinction matters, ask briefly or choose the language already dominant in the thread.
Each question should:
- present 2-3 strong options, not a blank request
- use concrete tradeoffs so the user can choose quickly
- help align on intent, boundaries, and autonomy level
- avoid jargon unless the user already introduced it
- stay concise, but not so short that the options become vague
Prefer two alignment rounds. Use one round only if the user already provided unusually complete requirements.
Entry Workflow
The first goal-todo pass should clarify three axes.
1. Goal Shape
Ship a Change (Recommended)The user wants a concrete delivered result such as code, docs, workflow, prompt system, or automation.Research and DecideThe user needs evidence, tradeoff analysis, or a decision memo before implementation.Open ExplorationThe user wants Codex to explore, structure, and progressively narrow a larger ambiguous problem.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2mo ago First seen · 177 lines · 73 tokens per session scan A 284d5b702486
goal-todo is a skill published in the GitHub repository Cooper-X-Oak/goal-to-do (3 stars, last pushed 5mo ago), licensed MIT. It adds 73 tokens to every session and 1,444 once invoked, about $0.0003 per session on Opus 5.5. 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 skills, from other repositories
gop
Use when a user wants a rough product, assistant, agent, chatbot, workflow, or business requirement converted into a visible expert-grade XML system prompt and immediately executed without waiting for approval.
anti-ai-slop-design
Remove AI slop and refine distinctive visual design in websites, landing pages, dashboards, and apps. Use for "less AI", "too generic", "premium polish", art direction, or decorative clutter: colored dots, card accent lines, border beams, icon bubbles, gradient blobs, and repetitive feature grids. Establish…
ui-ux
Design and meticulously style screens, components, dashboards, forms, and product flows. Use for UI/UX, "make it look better", visual refinement, perfectionist polish, typography, spacing, responsive layouts, design tokens, interaction states, accessibility, or developer handoff. Emphasize composition, optical…
premium-web-motion
Design, implement, and debug refined web animation, transitions, gestures, and micro-interactions. Use for motion polish, springs, interruption-safe dialogs, shared-element transitions, scroll effects, reduced motion, or animation performance. Keep motion purposeful, controllable, accessible, and measured; avoid…
security
Audit, threat-model, and harden application or AI-agent trust boundaries. Use for security reviews, authorization flaws, injection, SSRF, sessions, sensitive data, cryptography, webhooks, race conditions, dependency risks, or agent tool permissions. Trace actual exploit paths, prioritize evidence, and verify scoped…
web-development
Build, refactor, debug, and optimize web applications, frontend components, API routes, server actions, databases, and webhooks. Use for full-stack implementation, hydration, routing, caching, performance, or resilient mutations. Preserve the installed stack, validate trust boundaries, finish visual UI styling, and…