From 89c588db754b474b1426e21445d297d2951b99e9 Mon Sep 17 00:00:00 2001 From: Jammy2211 Date: Tue, 23 Jun 2026 10:12:14 +0100 Subject: [PATCH 1/2] docs: consolidate PyAutoFit agent instructions into canonical AGENTS.md Reduce .github/copilot-instructions.md to a thin pointer to AGENTS.md, the single source of truth (CLAUDE.md already @-imports it). AGENTS.md already covers the architecture and conventions more completely than the Copilot file, so nothing needed porting. Co-Authored-By: Claude Opus 4.8 (1M context) --- .github/copilot-instructions.md | 31 ++++++++----------------------- 1 file changed, 8 insertions(+), 23 deletions(-) diff --git a/.github/copilot-instructions.md b/.github/copilot-instructions.md index cb7dc5b34..350e7af3a 100644 --- a/.github/copilot-instructions.md +++ b/.github/copilot-instructions.md @@ -1,25 +1,10 @@ # Copilot Coding Agent Instructions -You are working on **PyAutoFit**, a Python probabilistic programming library for model fitting and Bayesian inference. - -## Key Rules - -- Run tests after every change: `python -m pytest test_autofit/` -- Format code with `black autofit/` -- All files must use Unix line endings (LF, `\n`) -- If changing public API (function signatures, class names, import paths), clearly document what changed in your PR description — downstream packages depend on this - -## Architecture - -- `autofit/non_linear/search/` — Non-linear search algorithms (MCMC, nested sampling, MLE) -- `autofit/mapper/` — Model composition and prior machinery -- `autofit/graphical/` — Graphical models and expectation propagation -- `autofit/aggregator/` — Results aggregation -- `autofit/database/` — SQLAlchemy results database -- `test_autofit/` — Test suite - -## Sandboxed runs - -```bash -NUMBA_CACHE_DIR=/tmp/numba_cache MPLCONFIGDIR=/tmp/matplotlib python -m pytest test_autofit/ -``` +You are working on **PyAutoFit**, a probabilistic programming library for model +composition and non-linear search / Bayesian inference (package `autofit`). + +The canonical, agent-agnostic instructions for this repo live in **`AGENTS.md`** at +the repository root — build and test commands, architecture, the model/search +subsystems, and conventions (including the `[nss]` / blackjax install gotcha). Read +it directly (Copilot does not process `@`-imports) and treat it as the single source +of truth. From c9a07e4637dabd75508aec15dda146153018572c Mon Sep 17 00:00:00 2001 From: Jammy2211 Date: Tue, 23 Jun 2026 10:12:51 +0100 Subject: [PATCH 2/2] docs: add signpost llms.txt routing to docs/ecosystem MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 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 to the API, the docs, AGENTS.md, and its place in the ecosystem. Co-Authored-By: Claude Opus 4.8 (1M context) --- llms.txt | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) create mode 100644 llms.txt diff --git a/llms.txt b/llms.txt new file mode 100644 index 000000000..4e7ab6e73 --- /dev/null +++ b/llms.txt @@ -0,0 +1,20 @@ +# PyAutoFit + +> PyAutoFit (package `autofit`) is a probabilistic programming library for model composition and non-linear search / Bayesian inference — `af.Model`/`af.Collection`, the `af.Analysis` interface, MCMC / nested-sampling / MLE searches, and a results aggregator and database. This file is a signpost: it points you to the right resource by intent — use the API, learn it, or work on the library itself. + +## Use it (examples & tutorials) + +- [autofit_workspace navigator](https://github.com/PyAutoLabs/autofit_workspace/blob/main/llms.txt): end-to-end example scripts and notebooks per task — the paste-friendly task router. Send any "how do I compose a model / run a search / inspect results?" question here. +- [HowToFit](https://github.com/PyAutoLabs/HowToFit): from-first-principles lecture series on model fitting and Bayesian inference. + +## API reference & docs + +- [PyAutoFit documentation (ReadTheDocs)](https://pyautofit.readthedocs.io/en/latest/): API reference, feature overviews, and the installation guide. + +## Work on it (contributors / coding agents) + +- [AGENTS.md](./AGENTS.md): build, tests, architecture, and conventions. + +## Ecosystem + +- Built on PyAutoConf (config); used by PyAutoGalaxy and PyAutoLens, which build `Analysis` subclasses on it.