Skip to content

docs: llms.txt, llms-full.txt and AGENTS.md - #68

Merged
petercorke merged 1 commit into
mainfrom
docs/agents-llms-txt
Oct 13, 2026
Merged

petercorke merged 1 commit into
mainfrom
docs/agents-llms-txt

Conversation

@petercorke

Copy link
Copy Markdown
Owner

Follows the pattern of ansitable and bdsim.

llms.txt and llms-full.txt

  • docs/llms_intro.md: hand-written, the part of llms.txt that an assistant that is using pgraph needs: what it is, a quick start, key points, common mistakes. Every claim in it was checked by running code (the quick start prints ['a', 'b', 'c'] 12.0; the planners return None for no path; component(c) takes a number; an in-place coordinate edit is not seen by the k-d tree; and so on).
  • docs/llms_txt.py: ansitable's generator, with one change: pgraph's toctree is nested (index → api → common → sections), so it follows toctrees recursively. llms.txt is the introduction followed by the list of pages; llms-full.txt is the Sphinx site rendered by the text builder, API reference included (13 pages, 175 kB, no raw directives or tracebacks).
  • docs/Makefile has a llms target, and the docs job of ci.yml runs make -C docs html llms, so both files are published with the site. Tried locally with exactly that command. The docs job runs only on pushes to main, so the files appear on the site after this merges.

AGENTS.md

Points assistants that are using pgraph to llms.txt, and records what an agent working on the source needs: the layout and the one Vertex class, the records that must agree (validate()), the two staleness flags and why they are separate, why Vertex.__setattr__ is hot and what writes __dict__ directly, pickling by a flat description, how tests and the scanners (Codacy, Codecov, mypy) behave, that performance claims are measured against 0.6.5, how a release is made, and that scipy must stay out of pyproject.toml. llms.txt links back to it under "Optional".

Also: a pointer to both files in the README.

🤖 Generated with Claude Code

llms.txt (llmstxt.org) is a short introduction for AI assistants: what pgraph is,
a quick start, key points and common mistakes, then links; llms-full.txt is
the whole Sphinx site as plain text, API reference included.  Both are built
by docs/llms_txt.py from the hand-written docs/llms_intro.md and the nested
toctrees (make html llms), and published with the site by the docs job.  The
introduction was checked claim by claim against the code.

AGENTS.md follows ansitable and bdsim: it points assistants that are using
pgraph to llms.txt, and records what an agent working on the source needs
to know: the layout, the records that must agree, the two staleness flags,
the hot Vertex.__setattr__, pickling, the tests and scanners, measuring
performance against 0.6.5, and how a release is made.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@codacy-production

Copy link
Copy Markdown

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Metrics 26 complexity · 0 duplication

Metric Results
Complexity 26
Duplication 0

View in Codacy

NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.

@petercorke
petercorke merged commit 7e763ea into main Oct 13, 2026
12 checks passed
@petercorke
petercorke deleted the docs/agents-llms-txt branch October 13, 2026 01:20
@codecov

codecov Bot commented Oct 13, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.26%. Comparing base (8210675) to head (0311897).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files
@@           Coverage Diff           @@
##             main      #68   +/-   ##
=======================================
  Coverage   97.26%   97.26%           
=======================================
  Files           3        3           
  Lines        1570     1570           
=======================================
  Hits         1527     1527           
  Misses         43       43           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

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