Skip to content

docs: add signpost llms.txt + consolidate agent instructions into AGENTS.md#570

Merged
Jammy2211 merged 2 commits into
mainfrom
feature/llms-txt-signpost
Jun 23, 2026
Merged

docs: add signpost llms.txt + consolidate agent instructions into AGENTS.md#570
Jammy2211 merged 2 commits into
mainfrom
feature/llms-txt-signpost

Conversation

@Jammy2211

Copy link
Copy Markdown
Collaborator

Summary

Adds a thin signpost llms.txt at the repo root and consolidates the remaining duplicated agent instructions into the canonical AGENTS.md.

llms.txt (new, 20 lines)

A front-door file following the llms.txt convention (one H1, a blockquote summary, then sections of one-line links). It only points — no source-tree catalogue, no duplication of AGENTS.md or the docs — routing an LLM by intent:

  • Use it → autolens_workspace navigator (llms.txt) + HowToLens
  • API & docs → ReadTheDocs
  • Work on the libraryAGENTS.md
  • Ecosystem → PyAutoGalaxy / PyAutoArray / PyAutoFit / PyAutoConf

All links verified live (PyAutoLabs org remotes; workspace llms.txt present on origin/main; ReadTheDocs URL confirmed from the README + .readthedocs.yaml).

.github/copilot-instructions.md (consolidated)

Replaced the duplicated Key Rules / Architecture / sandbox blocks with a thin pointer to AGENTS.md (the single source of truth, matching CLAUDE.md which already @-imports it). Also drops the slightly stale @to_array/@to_grid/@to_vector_yx decorator names in favour of AGENTS.md's accurate aa.decorators.* guidance.

Scripts Changed

None — docs only.

API Changes

None.

🤖 Generated with Claude Code

Jammy2211 and others added 2 commits June 23, 2026 09:00
Replace the duplicated Key Rules / Architecture / sandbox content in
.github/copilot-instructions.md with a thin pointer to AGENTS.md, the
single source of truth. CLAUDE.md already @-imports AGENTS.md; this brings
the Copilot instructions to the same convention and drops slightly stale
decorator names (@to_array/@to_grid/@to_vector_yx) in favour of AGENTS.md's
accurate aa.decorators.* guidance.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Thin front-door llms.txt at the repo root following the llms.txt
convention (H1 + blockquote summary, then sections of one-line links).
It only points — no source-tree catalogue, no AGENTS.md/docs duplication —
routing an LLM by intent: use the API (workspace navigator), learn lensing
(HowToLens), read the docs (ReadTheDocs), work on the library (AGENTS.md),
or follow the upstream ecosystem repos.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@Jammy2211 Jammy2211 merged commit 4b33d28 into main Jun 23, 2026
2 checks passed
@Jammy2211 Jammy2211 deleted the feature/llms-txt-signpost branch June 23, 2026 08:53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant