AnswerDotAI / AnswerDotAI/llms-txt

Proposal: complementary organizational discovery layer (ACI) for llms.txt

Open
#139 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
2.6k
Forks
163
Avg merge
1m
Merged PRs (30d)
1

Description

### ACI: A complementary organizational discovery layer for llms.txt

Hi there — I'm Eddie Wade, founder of Empire Labs and initial steward of the Autonomous Company Interface (ACI) draft.

**ACI is an open specification that allows autonomous agents to discover an organization's identity, capabilities, knowledge, trust assertions, and exposed agents through linked machine-readable manifests** (`.well-known/aci/`).

I'm reaching out because I believe ACI is designed to sit naturally alongside llms.txt:

- **llms.txt** describes what content a website offers to language models (the "what")
- **ACI** describes the organization behind it — its identity, capabilities, trust model, and agent surfaces (the "who" and "how")

We built ACI's discovery mechanism to be compatible with the llms.txt ecosystem. An agent discovering `llms.txt` could follow links into ACI manifests, and vice versa.

**We're holding the specification at v0.9 until we have three independent implementations.** Before promoting toward v1.0, I'd value your review — particularly whether the ACI discovery model complements your existing architecture or creates unnecessary overlap.

Key links:
- **Overview/Article:** https://dev.to/narko4u/the-missing-layer-for-autonomous-agents-introducing-the-autonomous-company-interface-aci-2d75
- **Specification:** https://github.com/narko4u/aci-spec
- **Quick-start template:** https://github.com/narko4u/aci-pages-template
- **Implementation tracker:** https://github.com/narko4u/aci-spec/issues/6

I'm not asking for endorsement. Honest feedback about anything unclear, unnecessary, or missing would be genuinely valuable.

Thanks for your time and for building llms.txt — it's been a direct inspiration for ACI's design.

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by reviewing the linked ACI overview and specification, then compare its discovery model with the llms.txt project described in this issue. There are no repository files or tests identified; done would require a maintainer decision or documented feedback on whether ACI complements the project.

Written by the indexing model from the issue text.

Assessment

Domain
documentation
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.