Shayan Rais
b5b56f65e5
Embeds two X.com screenshots (Alex Finn, Meta Alchemist) with verbatim tweet text and source URLs. Covers two patterns: asking the agent to propose its own /goals, and asking it to draft a /goal prompt from session + repo context. Co-Authored-By: Claude <noreply@anthropic.com>
87 lines
3.0 KiB
Markdown
87 lines
3.0 KiB
Markdown
# Goal Implementation
|
|
|
|
|
|
|
|
<table width="100%">
|
|
<tr>
|
|
<td><a href="../">← Back to Claude Code Best Practice</a></td>
|
|
<td align="right"><img src="../!/claude-jumping.svg" alt="Claude" width="60" /></td>
|
|
</tr>
|
|
</table>
|
|
|
|
---
|
|
|
|
<a href="#goal-tips-from-the-community"><img src="../!/tags/implemented-hd.svg" alt="Implemented"></a>
|
|
|
|
`/goal` keeps your agent working across turns until a condition is satisfied — Claude Code, Codex, and Hermes Agent all support it. The community is converging on a few high-leverage prompting tricks that pair well with it.
|
|
|
|
---
|
|
|
|
## Goal Tips from the Community
|
|
|
|
### 1. Ask the agent to propose its own goals
|
|
|
|
<p align="center">
|
|
<img src="assets/impl-goal-claude.png" alt="Alex Finn tweet — /goal is the most underrated AI feature of 2026" width="50%">
|
|
</p>
|
|
|
|
> It's official. Claude Code just released /goal
|
|
>
|
|
> The single most underrated AI feature of 2026
|
|
>
|
|
> Now Claude Code, Codex, and Hermes agent have it
|
|
>
|
|
> It allows your agent to complete long running tasks, sometimes for days
|
|
>
|
|
> EVERYONE should be immediately running this prompt:
|
|
>
|
|
> 'Based on what you know about me, my goals, ambitions, and what we've built together already, what are the 3 /goals we can run right now that would run for long time periods and produce the best results?'
|
|
>
|
|
> Choose one, then ask for it to build you a prompt
|
|
>
|
|
> You should get a few options for super powerful goal prompts that will have your agent of choice complete long running tasks that will deliver your mind blowing results.
|
|
>
|
|
> Carve out 15 minutes tonight to do this. Thank me later.
|
|
|
|
**Source:** [Alex Finn (@AlexFinn) on X](https://x.com/AlexFinn/status/2053976411296452887)
|
|
|
|
---
|
|
|
|
### 2. Let the agent draft the /goal prompt for you
|
|
|
|
<p align="center">
|
|
<img src="assets/impl-goal-codex.png" alt="Meta Alchemist tweet — /goal trick for Codex" width="50%">
|
|
</p>
|
|
|
|
> wanna know the best /goal trick for Codex?
|
|
>
|
|
> just tell your Codex:
|
|
>
|
|
> "read this session and repo, analyze deeply the exact intent and goals we are looking to achieve here then write me the /goal prompt for this.
|
|
>
|
|
> make sure to dig into history & docs we have to be 100% clear"
|
|
>
|
|
> also you can add:
|
|
>
|
|
> "if you are not sure about certain parts or wanna ask me a few questions to clarify certain goals further don't hesitate"
|
|
>
|
|
> then just copy paste what Codex gives you change the initial part to /goal
|
|
>
|
|
> and it will do exactly what you wanted to do in that session / repo, nonstop until it gets to completion.
|
|
|
|
**Source:** [Meta Alchemist (@meta_alchemist) on X](https://x.com/meta_alchemist/status/2054214497443995694)
|
|
|
|
---
|
|
|
|
## 
|
|
|
|
```bash
|
|
$ claude
|
|
> /goal <condition>
|
|
> /goal clear
|
|
```
|
|
|
|
`/goal <condition>` keeps Claude working across turns until a Haiku-evaluated condition holds. It is complementary to `/loop` (time-driven) and auto mode (per-tool). Requires Claude Code v2.1.139+.
|
|
|
|
See the [official docs](https://code.claude.com/docs/en/goal) for full behavior.
|