# AGENTS.MD doesnt seem to work reliably if at all

**URL:** <https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964>\
**Category:** Discussions\
**Created:** [January 4, 2026, 10:57pm UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964 "2026-01-04T22:57:07Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![Peter\_Cox](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/peter_cox/32/100910_2.png) [@Peter\_Cox](https://forum.cursor.com/u/Peter_Cox)\
**Post date:** [January 4, 2026, 10:57pm UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/1 "2026-01-04T22:57:07Z")

</div>

I can never get my AGENTS.MD to be reliably followed.

A good example is the useMemo or useCallback. It frequently loves to add those anyway on more vibe-codey parts of my code where I’m experimenting unless I have a very solid architectural pattern defined.

 ![image](https://us1.discourse-cdn.com/cursor1/original/3X/2/7/2723344af64d293a3d3a9f53bb56b0aa4aadde42.png)

 ![image](https://us1.discourse-cdn.com/cursor1/original/3X/c/2/c2d26dcb914a8eff2673c9d26f928da877cc3fcb.jpeg)

---

<div class="post-metadata">

**Author:** ![liquefy](https://avatars.discourse-cdn.com/v4/letter/l/edb3f5/32.png) [@liquefy](https://forum.cursor.com/u/liquefy)\
**Post date:** [January 4, 2026, 11:13pm UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/2 "2026-01-04T23:13:22Z")

</div>

> [@How do I get agents to actually listen to me?](https://forum.cursor.com/t/how-do-i-get-agents-to-actually-listen-to-me/147130/3):
>
> I asked agent to rewrite rules so it understands them better and it worked, saw other people mention this on forum too.

more context used = less reliable to follow rules

too long rules = bad

too many tools/mcp servers = bad

---

<div class="post-metadata">

**Author:** ![Peter\_Cox](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/peter_cox/32/100910_2.png) [@Peter\_Cox](https://forum.cursor.com/u/Peter_Cox)\
**Post date:** [January 4, 2026, 11:23pm UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/3 "2026-01-04T23:23:13Z")

</div>

Thanks for the advice - I’ve noticed I’m on about 1000 lines now for AGENTS.MD which I generally let it do its own thing so will give that a trim.

I don’t have any MCP tools though and not convinced on the context as it happens early into a chat

---

<div class="post-metadata">

**Author:** ![Naufaldi\_Rafif](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/naufaldi_rafif/32/73623_2.png) [@Naufaldi\_Rafif](https://forum.cursor.com/u/Naufaldi_Rafif)\
**Post date:** [January 5, 2026, 4:52am UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/4 "2026-01-05T04:52:52Z")

</div>

1,000 lines for an AGENTS.md is a lot! You can read this blog post—a typical AGENTS.md is only about 300 lines. Are you working on a monorepo? Why is that file so long?

- [https://www.humanlayer.dev/blog/writing-a-good-claude-md](https://www.humanlayer.dev/blog/writing-a-good-claude-md)

If you need a code guide, it is better to use ESLint and add these rules:

- [https://www.eslint-react.xyz/docs/rules/no-unnecessary-use-callback](https://www.eslint-react.xyz/docs/rules/no-unnecessary-use-callback)
- [https://www.eslint-react.xyz/docs/rules/no-unnecessary-use-memo](https://www.eslint-react.xyz/docs/rules/no-unnecessary-use-memo)

Then, in your instructions, ask the AI to always run “lint --fix” after finishing a task. If a warning or error shows up, the model will try to fix it. Don’t add unnecessary code styles to agents; use a linter for that. Or add rules and apply them intelligently or manually.

---

<div class="post-metadata">

**Author:** ![Peter\_Cox](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/peter_cox/32/100910_2.png) [@Peter\_Cox](https://forum.cursor.com/u/Peter_Cox)\
**Post date:** [January 6, 2026, 11:12pm UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/5 "2026-01-06T23:12:28Z")

</div>

All good suggestions thanks.

I trimmed my AGENTS.MD to circa 99 lines.

Using Composer-1 I still get this crap to support made-up backwards compatibility despite line “Never support backwards compatibility.

It still peppers XML everywhere despite line 14 saying this.

“Fake news” is how I sum up these stupid files

 ![image](https://us1.discourse-cdn.com/cursor1/original/3X/c/e/ceef9dd42d4e429f9c39eaad02b9e83957b551f5.png)

 ![image](https://us1.discourse-cdn.com/cursor1/original/3X/9/d/9de75b54fb3d5a46c1b8b1c24a15c51baa204080.png)

---

<div class="post-metadata">

**Author:** ![Naufaldi\_Rafif](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/naufaldi_rafif/32/73623_2.png) [@Naufaldi\_Rafif](https://forum.cursor.com/u/Naufaldi_Rafif)\
**Post date:** [January 7, 2026, 1:45am UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/6 "2026-01-07T01:45:26Z")

</div>

Nice, always try to plan first befoe work on anything. even for small task. use Cursor Plan since its get better for its. Here example of my plan, so i know excactly what Models will do, you can make command /plan for easy make plan

 ![Screenshot 2026-01-07 at 08.44.20](https://us1.discourse-cdn.com/cursor1/original/3X/b/9/b9f73ddfd27f50a57fa57ced5bac508ec7c7eec4.jpeg)

my command /plan

 ![Screenshot 2026-01-07 at 08.45.16](https://us1.discourse-cdn.com/cursor1/original/3X/a/a/aa7b9cca36733984c6db3e26d41ad0afe357b1cc.png)

---

<div class="post-metadata">

**Author:** ![system](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/system/32/99069_2.png) [@system](https://forum.cursor.com/u/system)\
**Post date:** [April 7, 2026, 1:45am UTC](https://forum.cursor.com/t/agents-md-doesnt-seem-to-work-reliably-if-at-all/147964/7 "2026-04-07T01:45:52Z")

</div>

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
