RoyLee1224 opened a new pull request, #68204:
URL: https://github.com/apache/airflow/pull/68204

   related: #62500 
   
   ## Problem
   AGENTS.md (468 lines) is loaded in full at session start. Anthropic's own 
[guidance](https://code.claude.com/docs/en/memory#write-effective-instructions) 
recommends keeping CLAUDE.md **under 200 lines** — longer files consume more 
context and reduce adherence. In long multi-turn sessions, agents accumulate 
enough context to enter the "Dumb Zone": they revert to generic Python patterns 
and lose Airflow-specific routing (e.g. choosing uv over breeze for a 
db_test-only case).
   
   A prerequisite to addressing this (via modular skills, selective context 
loading) is having a single source of truth for the commands that AGENTS.md 
exposes. Today the Commands section is hand-maintained independently from the 
contributing docs — they can drift without anyone noticing.
   
   ## What this PR does
   Makes contributing-docs the single source of truth for the AGENTS.md 
Commands section (#62500).
   - Adds `.. AGENT-SKILL-START` / `.. AGENT-SKILL-END` markers to three 
contributing docs (08_static_code_checks.rst, testing/unit_tests.rst, 
05_pull_requests.rst)
   - Adds a generate-agent-skills prek hook that extracts them and syncs the 
Commands block in AGENTS.md via `insert_documentation()`
   - The existing prek CI job catches drift automatically
   - AGENTS.md content is unchanged — the only diff is two HTML comment markers 
wrapping the existing commands
   
   ## Future Work
   With contributing-docs as the enforced source of truth, subsequent PRs can:
   1. **Modular skills** — move commands from AGENTS.md into on-demand SKILL.md 
files so agents load only what's relevant to the current task, reducing context 
pressure
   2. **AGENTS.md slimdown** — keep only judgment rules (architecture 
boundaries, coding standards) in the always-loaded file, targeting the 
recommended < 200 lines
   3. **Evaluation harness** — context-pressure benchmarks measuring skill lift 
(with-skill vs without-skill baseline) and routing accuracy (correct 
breeze/uv/--project selection) over long sessions, to validate that future 
skill changes actually improve agent compliance
   
    <!-- SPDX-License-Identifier: Apache-2.0
         https://www.apache.org/licenses/LICENSE-2.0 -->
   
   <!--
   Thank you for contributing!
   
   Please provide above a brief description of the changes made in this pull 
request.
   Write a good git commit message following this guide: 
http://chris.beams.io/posts/git-commit/
   
   Please make sure that your code changes are covered with tests.
   And in case of new features or big changes remember to adjust the 
documentation.
   
   Feel free to ping (in general) for the review if you do not see reaction for 
a few days
   (72 Hours is the minimum reaction time you can expect from volunteers) - we 
sometimes miss notifications.
   
   In case of an existing issue, reference it using one of the following:
   
   * closes: #ISSUE
   * related: #ISSUE
   -->
   
   ---
   
   ##### Was generative AI tooling used to co-author this PR?
   
   <!--
   If generative AI tooling has been used in the process of authoring this PR, 
please
   change below checkbox to `[X]` followed by the name of the tool, uncomment 
the "Generated-by".
   -->
   
   - [X] Yes — Claude Code (Opus 4.6)
   
   <!--
   Generated-by: [Tool Name] following [the 
guidelines](https://github.com/apache/airflow/blob/main/contributing-docs/05_pull_requests.rst#gen-ai-assisted-contributions)
   -->
   
   ---
   
   * Read the **[Pull Request 
Guidelines](https://github.com/apache/airflow/blob/main/contributing-docs/05_pull_requests.rst#pull-request-guidelines)**
 for more information. Note: commit author/co-author name and email in commits 
become permanently public when merged.
   * For fundamental code changes, an Airflow Improvement Proposal 
([AIP](https://cwiki.apache.org/confluence/display/AIRFLOW/Airflow+Improvement+Proposals))
 is needed.
   * When adding dependency, check compliance with the [ASF 3rd Party License 
Policy](https://www.apache.org/legal/resolved.html#category-x).
   * For significant user-facing changes create newsfragment: 
`{pr_number}.significant.rst`, in 
[airflow-core/newsfragments](https://github.com/apache/airflow/tree/main/airflow-core/newsfragments).
 You can add this file in a follow-up commit after the PR is created so you 
know the PR number.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to