)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"047c79315a660d5e48e7cfef2d2ae4814c6791a3","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"67663a0b_71ceaeae","updated":"2026-05-01 18:20:55.000000000","message":"I don\u0027t think this needs to be a community goal. First, not everyone will want the stripped-down AGENTS.md approach that nova currently does (and which is codified in this document). Second, I don\u0027t think that all projects _need_ such a thing and that tracking this as a community goal sort of implies a bunch of process and accounting about completion that just doesn\u0027t matter.\n\nHaving a cross-project working group on determining best practices and collaborating on a common repo is totally fine, and a good idea. I just think the community goal stick is too large and heavyweight a lever at the moment. To me, this is a tool to push things through that must be done. Things like RBAC, FIPS, eventlet removal, etc. Standardizing an AGENTS.md in each repo is not something we _need_ to do, IMHO.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":10342,"name":"Jay Faulkner","display_name":"JayF","email":"jay@jvf.cc","username":"JayF","status":"youtube.com/@oss-gr / podcast.gr-oss.io"},"change_message_id":"006f51934afa16189f4acb929ab747e80daaf981","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"34ebacf4_e475d334","updated":"2026-04-30 14:50:14.000000000","message":"I would strongly prefer this be handed at the per-project level. We are extremely likely to add one -- but I don\u0027t think we should, at a TC level, enforcing cooperation with or use of LLM tooling.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"8498735eb157b17634a0b12c441079f825d14710","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"5d4a5819_95f117d9","updated":"2026-05-05 16:27:56.000000000","message":"Thanks all for the reviews, we\u0027ll discuss those concerns hopefully at today\u0027s TC meeting (5pm UTC) and we have a consensus about it, I\u0027ll abandon this change.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"ab80f142_008f14ba","updated":"2026-05-03 05:50:57.000000000","message":"Thanks for putting this together, Sylvain. The underlying problem is clear that AI tools get OpenStack\u0027s workflow wrong because our contributor conventions (Gerrit, stestr, hacking/flake8, DCO) are arguably \"unusual\" in the open-source world. \nWe\u0027ve tried tackling this problem for human contributors through extensive contributor documentation. We\u0027ve probably faltered a bit over time and let tribal knowledge creep in. \n\nI would support a continuous effort to audit and correct issues in the contributor guides; and together as a community, we check on this regularly and get better over time. Such a thing will help humans and AI tools; and, your proposal needs it too.. really, because without good docs, the AGENTS.md is not good either.\n\nI agree with the others here. I don\u0027t think a community goal is the right vehicle for this, and I don\u0027t think the approach is mature enough to mandate across our nearly 600 code repositories. A few concerns:\n\nCommunity goals are for things that must be done. As Dan noted, community goals carry real weight; we want PTLs to formally commit, the TC tracks the goal completion, teams must prioritize this work. That mechanism is indeed appropriate for eventlet removal, RBAC, or even FIPS which we didn\u0027t end up keeping. AGENTS.md is a convenience optimization for contributors who choose to use AI tools. It\u0027s a good idea for projects that want it, but it\u0027s not something the TC needs to mandate.\n\nAs I\u0027ve noted inline, the format isn\u0027t mature enough to standardize on. There are fundamental gaps in the \"open standard\" that should be resolved before we require it across our repositories. Else, we risk annoying and alienating our contributors. \n\nAnd, finally my concern with all of this is that this space evolves so quickly and transforms every couple of months with a new model/harness/agentic tool drop. So we could become quite busy in catering to the mess instead of solving our mission.\n\nWhat I\u0027d suggest instead:\n\n  1. Projects that want to adopt AGENTS.md should do so. As you noted, Nova and Neutron are doing great work here and I encourage them to continue experimenting so we can learn how that evolves. We don\u0027t stop anyone else from attempting this either. \n  2. The TC can issue a resolution/recommendation encouraging projects to make their contributor documentation more consumable for humans AND machines.\n  3. After 1-2 cycles of organic adoption and learning, if the approach proves broadly valuable AND the format stabilizes, we can see whether a community goal is warranted.\n\nDoing this will avoid spending TC governance capital on tracking whether every repo has a markdown file.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"28bb939e005e10b9a7f73abd95d7b845ee3a457a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"626c2380_3ec43c0c","updated":"2026-05-04 20:26:56.000000000","message":"Thanks sylvain for the start, definitely this will be needed soon but I also do not think we are in that stage where we need to define it as a goal. I prefer to progress it as a popup team or working group and provide more documents/preferred way in project-team-guide and let project evolve those as per their need.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":10342,"name":"Jay Faulkner","display_name":"JayF","email":"jay@jvf.cc","username":"JayF","status":"youtube.com/@oss-gr / podcast.gr-oss.io"},"change_message_id":"cce72ff2ed7e1b1a3d9c10156bfe05a42ae97121","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"db77bf0e_2603349b","in_reply_to":"34ebacf4_e475d334","updated":"2026-04-30 14:50:36.000000000","message":"We\u003dIronic, to be clear","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"821a14dcba0e69273fb571d2cbbdffef8ec885fe","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"0bc0b203_b7334774","in_reply_to":"ab80f142_008f14ba","updated":"2026-05-04 10:19:23.000000000","message":"Lot of valid concerns listed here. I we go with a recommendation instead of a goal. Can this be extended to cover the usage of the common openstack/agentic-workflows repository? I have the feeling that such common repo can only serve the needs of the community if we can define / agree what is the common goal that repo solves.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"d39f642f52ad25667f39247f66dabff92aeaeef9","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"1d6caf9b_009fc151","in_reply_to":"db77bf0e_2603349b","updated":"2026-04-30 15:52:21.000000000","message":"I think wider cooperation is needed than per project. This is visible from the effort started in openstack/agentic-workflows to collect common reusable skills personas for the *whole* openstack namespace and I guess individual project level AGENTS.md files should reference the common stuff instead of duplicating them.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"}],"goals/proposed/agents-md.rst":[{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":9,"context_line":"explicit project-level guidance, these agents default to wrong tools (pytest"},{"line_number":10,"context_line":"instead of stestr, ruff instead of hacking/flake8), miss OpenStack-specific"},{"line_number":11,"context_line":"workflows (Gerrit instead of GitHub PRs, tox environments, DCO sign-off), and"},{"line_number":12,"context_line":"hallucinate conventions that do not exist."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"A terse ``AGENTS.md`` file at the root of each project repository solves this"},{"line_number":15,"context_line":"by acting as a routing index that directs agents to the project\u0027s **existing**"}],"source_content_type":"text/x-rst","patch_set":1,"id":"45b40e02_7b996147","line":12,"range":{"start_line":12,"start_character":41,"end_line":12,"end_character":42},"updated":"2026-05-03 05:50:57.000000000","message":"A mechanics question on this broad effort:\n\nDoesn\u0027t everyone, when operating an AI tool on a repository, perform an \"init\"? (e.g.: [1][2][3]). So presumably they already customized it to their heart\u0027s content. No tools provide a way, in my knowledge, to override AGENTS.md with a local flavor; a git rebase/pull will end up clobbering what the developer had. \n\nSo, in my humble view, integrating at this level in nearly 600 repos will frustrate developers imho. The format hasn\u0027t matured enough to handle the most basic\nmulti-user scenario. This is probably okay for now in many projects outside OpenStack - but, let\u0027s consider this please before we wield a sledgehammer. \n\n\n[1] https://code.claude.com/docs/en/best-practices#write-an-effective-claude-md\n[2] https://developers.openai.com/codex/learn/best-practices#make-guidance-reusable-with-agentsmd\n[3] https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference#command-line-commands","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":11,"context_line":"workflows (Gerrit instead of GitHub PRs, tox environments, DCO sign-off), and"},{"line_number":12,"context_line":"hallucinate conventions that do not exist."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"A terse ``AGENTS.md`` file at the root of each project repository solves this"},{"line_number":15,"context_line":"by acting as a routing index that directs agents to the project\u0027s **existing**"},{"line_number":16,"context_line":"authoritative documentation. The file is intentionally small (under 100 lines)"},{"line_number":17,"context_line":"and contains no duplicated rules — it points to contributor guides, test"},{"line_number":18,"context_line":"commands, and architecture docs that already live in-tree."}],"source_content_type":"text/x-rst","patch_set":1,"id":"f17fa4ea_143a6693","line":15,"range":{"start_line":14,"start_character":66,"end_line":15,"end_character":28},"updated":"2026-05-03 05:50:57.000000000","message":"this is a strong claim, from what i can tell, the AAIF defined this standard, but, today\u0027s implementation isn\u0027t universal.. it\u0027s probably coming, or, who knows, the spec could change as well.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":26,"context_line":"tools (Claude Code, Cursor, GitHub Copilot, and others). It uses plain Markdown"},{"line_number":27,"context_line":"with no required schema, making it trivial to author and maintain."},{"line_number":28,"context_line":""},{"line_number":29,"context_line":"This goal is fully compliant with the `OpenInfra AI policy`_ adopted in June"},{"line_number":30,"context_line":"2024, which requires human-in-the-loop review and ``Assisted-By:`` commit"},{"line_number":31,"context_line":"labels for AI-assisted contributions. The ``AGENTS.md`` file itself does not"},{"line_number":32,"context_line":"generate code — it only provides context to agents that contributors choose to"},{"line_number":33,"context_line":"use."}],"source_content_type":"text/x-rst","patch_set":1,"id":"3ddaecc8_219fae7d","line":30,"range":{"start_line":29,"start_character":0,"end_line":30,"end_character":4},"updated":"2026-05-03 05:50:57.000000000","message":"Adding agents.md has little to do with commit level annotations, so I am not sure why this was called out. Are you proposing a section in AGENTS.md to handle this commit annotation?\n\n\nnote: I think we, the TC, needs to first acknowledge this policy and cross link it in OpenStack documentation.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":39,"context_line":"The file should contain:"},{"line_number":40,"context_line":""},{"line_number":41,"context_line":"#. **Project identity**: name, brief description, primary programming language."},{"line_number":42,"context_line":"#. **Build \u0026 test commands**: exact ``tox`` environments and flags"},{"line_number":43,"context_line":"   (e.g. ``tox -e py313``, ``tox -e pep8``), the test runner"},{"line_number":44,"context_line":"   (``stestr`` not ``pytest``), and any project-specific test patterns."},{"line_number":45,"context_line":"#. **Code style tools**: which linter enforces style (``hacking``/``flake8``,"},{"line_number":46,"context_line":"   not ``ruff``), any project-specific hacking checks."}],"source_content_type":"text/x-rst","patch_set":1,"id":"aa9f02a9_2b8d0a7b","line":43,"range":{"start_line":42,"start_character":29,"end_line":43,"end_character":60},"updated":"2026-05-03 05:50:57.000000000","message":"why this duplication?\n\nisn\u0027t that what tox.ini is for? can\u0027t we add a redirect to that? or, point at contributor docs explaining how testing is done, and what testing is expected?","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":41,"context_line":"#. **Project identity**: name, brief description, primary programming language."},{"line_number":42,"context_line":"#. **Build \u0026 test commands**: exact ``tox`` environments and flags"},{"line_number":43,"context_line":"   (e.g. ``tox -e py313``, ``tox -e pep8``), the test runner"},{"line_number":44,"context_line":"   (``stestr`` not ``pytest``), and any project-specific test patterns."},{"line_number":45,"context_line":"#. **Code style tools**: which linter enforces style (``hacking``/``flake8``,"},{"line_number":46,"context_line":"   not ``ruff``), any project-specific hacking checks."},{"line_number":47,"context_line":"#. **Architecture overview pointer**: a one-line description with a path to"},{"line_number":48,"context_line":"   the in-tree architecture doc (e.g. ``doc/source/contributor/``)."},{"line_number":49,"context_line":"#. **Contribution workflow**: Gerrit-based review (not GitHub PRs), commit"}],"source_content_type":"text/x-rst","patch_set":1,"id":"634eae76_1ecd11f8","line":46,"range":{"start_line":44,"start_character":3,"end_line":46,"end_character":54},"updated":"2026-05-03 05:50:57.000000000","message":"All of this is in the contributor docs that we should be redirecting to?\n\nOr is there an efficiency thingy that i\u0027m missing?","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":47,"context_line":"#. **Architecture overview pointer**: a one-line description with a path to"},{"line_number":48,"context_line":"   the in-tree architecture doc (e.g. ``doc/source/contributor/``)."},{"line_number":49,"context_line":"#. **Contribution workflow**: Gerrit-based review (not GitHub PRs), commit"},{"line_number":50,"context_line":"   message conventions, ``git-review`` usage, and sign-off requirements."},{"line_number":51,"context_line":"#. **Links to existing docs**: paths to contributor guide, API reference,"},{"line_number":52,"context_line":"   release notes, and any other relevant documentation."},{"line_number":53,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"13ecc9cc_646cb033","line":50,"range":{"start_line":50,"start_character":50,"end_line":50,"end_character":71},"updated":"2026-05-03 05:50:57.000000000","message":"This is where this can get quite problematic. Should AGENTS.md offer git commit and review instructions? In my view, many people\u0027d hate that and want to explicitly disable this sort of thing - they may want complete control of commit and \"push\" workflows, and may not trust agents to handle this.\n\nAs a dire example, over the last few weeks, code pertaining to a few embargoed security vulnerabilities were accidentally pushed out to gerrit by enthusiastic \"agents\". This is a bad pattern, and what silently suffers is our community\u0027s processes and reputation.\n\nPlease, let\u0027s not require this in the goal. Humans can do as they please, but let the community set secure standards.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":54,"context_line":"Key principles:"},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"- **No duplication.** ``AGENTS.md`` references in-tree docs, it does not"},{"line_number":57,"context_line":"  inline rules. If a convention is already documented in"},{"line_number":58,"context_line":"  ``doc/source/contributor/code-review.rst``, the file points there instead"},{"line_number":59,"context_line":"  of restating the rules. This prevents drift between the routing file and"},{"line_number":60,"context_line":"  the canonical documentation."}],"source_content_type":"text/x-rst","patch_set":1,"id":"c8ec6b2d_797ff035","line":57,"range":{"start_line":57,"start_character":2,"end_line":57,"end_character":14},"updated":"2026-05-03 05:50:57.000000000","message":"the testing stuff, the test runner stuff, the git commit stuff and the review stuff above contradicts this","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"4423264bd3ac76201b16c8cdaad2be3080e97094","unresolved":true,"context_lines":[{"line_number":62,"context_line":"  add AI skills, agent personas, or any other agentic tooling to their"},{"line_number":63,"context_line":"  repositories. Only the routing file is in scope."},{"line_number":64,"context_line":"- **No CI or code changes.** The file is pure Markdown documentation. It does"},{"line_number":65,"context_line":"  not affect builds, tests, or packaging."},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"Optional additions that projects may choose to include:"},{"line_number":68,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"ea006141_fdf79f3b","line":65,"updated":"2026-04-30 15:50:22.000000000","message":"Do we want a markdown syntax check on the file?\nMaybe a check that the links in that file point to existing documents?\n\nSure the usefulness of the AGENTS.md file is hard to verify in CI today.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"4423264bd3ac76201b16c8cdaad2be3080e97094","unresolved":true,"context_lines":[{"line_number":71,"context_line":"- An ``agentic-coding.rst`` contributor doc explaining the project\u0027s"},{"line_number":72,"context_line":"  conventions for AI-assisted contributions, ``Assisted-By:`` label usage,"},{"line_number":73,"context_line":"  and how to maintain ``AGENTS.md``."},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"Goal Checklist"},{"line_number":76,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":77,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"b18903c1_7fd75d81","line":74,"updated":"2026-04-30 15:50:22.000000000","message":"How this relates to the effort being done in the openstack/agentic-workflows ?","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":80,"context_line":""},{"line_number":81,"context_line":"Status: NO"},{"line_number":82,"context_line":""},{"line_number":83,"context_line":"The direction was agreed upon at the Nova 2026.2 Hibiscus PTG: a terse routing"},{"line_number":84,"context_line":"file that points to existing docs rather than inlining rules. A draft proposal"},{"line_number":85,"context_line":"for Nova is under review at https://review.opendev.org/c/openstack/nova/+/984540"},{"line_number":86,"context_line":"but the exact content and structure of ``AGENTS.md`` are not yet finalised."}],"source_content_type":"text/x-rst","patch_set":1,"id":"f17e778d_cb5d618d","line":83,"range":{"start_line":83,"start_character":4,"end_line":83,"end_character":61},"updated":"2026-05-03 05:50:57.000000000","message":"ack, as others have noted this experiment can happen in nova, and we can derive an understanding of this before we push this out to the community and mandate across all OpenStack repos.\n\nThe intent of the TC\u0027s questions for this checklist is to evaluate readiness. If the answer is No, usually, the goal isn\u0027t accepted. Was your intent to transition these to YES before we accept the goal?","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":119,"context_line":"  (community repository for shared skills, personas, and validation tooling)"},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"The approach was discussed and agreed upon at the Nova 2026.2 Hibiscus PTG"},{"line_number":122,"context_line":"session. The key consensus point was that project-specific knowledge belongs"},{"line_number":123,"context_line":"in each project\u0027s own repository (not centralised). If the community agrees"},{"line_number":124,"context_line":"that shared cross-project skills and personas would be useful, a common home"},{"line_number":125,"context_line":"could be established for them; an experimental proposal already exists at"},{"line_number":126,"context_line":"``openstack/agentic-workflows``, but this is not yet an endorsed location."}],"source_content_type":"text/x-rst","patch_set":1,"id":"66bfb190_6b4b17a2","line":123,"range":{"start_line":122,"start_character":9,"end_line":123,"end_character":51},"updated":"2026-05-03 05:50:57.000000000","message":"Do you want to make the point here that Nova contributors can speak to the community of ~40+ project teams :) As other comments have alluded, i think can be a very contentious direction, and as the TC, we should drive and drive consensus across the community, not cargo cult decisions.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"4423264bd3ac76201b16c8cdaad2be3080e97094","unresolved":true,"context_lines":[{"line_number":131,"context_line":"Discussion happens on the ``openstack-discuss`` mailing list and primarily in"},{"line_number":132,"context_line":"the ``#openstack-tc`` IRC channel. Each project should also use its own IRC"},{"line_number":133,"context_line":"channel for discussion of its related patches. As more projects adopt the goal,"},{"line_number":134,"context_line":"coordination can move to a dedicated etherpad or IRC meeting if needed."},{"line_number":135,"context_line":""},{"line_number":136,"context_line":"Completion Date \u0026 Criteria"},{"line_number":137,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"58290ca6_6657b62c","line":134,"updated":"2026-04-30 15:50:22.000000000","message":"There is a room called OFTC/#openstack-agentic-workflows does it something we use for this work?","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":138,"context_line":""},{"line_number":139,"context_line":"Milestone 1: 2027.2"},{"line_number":140,"context_line":""},{"line_number":141,"context_line":"#. Every official OpenStack project deliverable has an ``AGENTS.md`` file at"},{"line_number":142,"context_line":"   the root of its primary repository."},{"line_number":143,"context_line":"#. The file references the project\u0027s actual build commands, test runner,"},{"line_number":144,"context_line":"   style enforcement tools, and contribution documentation."},{"line_number":145,"context_line":"#. The file does NOT duplicate rules already enforced by CI (linters, hacking"}],"source_content_type":"text/x-rst","patch_set":1,"id":"478b4baf_c2f236ac","line":142,"range":{"start_line":141,"start_character":3,"end_line":142,"end_character":38},"updated":"2026-05-03 05:50:57.000000000","message":"Our governance (projects.yaml) currently lists 40+ project teams governing nearly 600 repositories. This criterion requires every one of those repos to add and maintain an AGENTS.md file by 2027.2?\n\nSounds like the wrong priority. \n\nA more proportionate criterion would be \"projects that choose to adopt AGENTS.md follow a common structure\" rather than \"every deliverable must have one.\"","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":144,"context_line":"   style enforcement tools, and contribution documentation."},{"line_number":145,"context_line":"#. The file does NOT duplicate rules already enforced by CI (linters, hacking"},{"line_number":146,"context_line":"   checks, etc.) — it points to the docs that describe those rules."},{"line_number":147,"context_line":"#. The file follows the `open standard`_ Markdown format and is usable by"},{"line_number":148,"context_line":"   any compatible AI coding tool."},{"line_number":149,"context_line":""},{"line_number":150,"context_line":"Champion"},{"line_number":151,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"25620385_cfc68ea3","line":148,"range":{"start_line":147,"start_character":3,"end_line":148,"end_character":33},"updated":"2026-05-03 05:50:57.000000000","message":"This is repetition, but, i\u0027ll say it here too, the standard is evolving, and without a way for us to test every AI tool (or atleast a bunch of them) in the CI, we might be opening the doors for a bunch of weird things that work for some, break for others - and we\u0027ll end up manually policing edits and causing a needless churn.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":16643,"name":"Goutham Pacha Ravi","email":"gouthampravi@gmail.com","username":"gouthamr"},"change_message_id":"6ca80987c55b5a441cae1c991ab91fd89edb796f","unresolved":true,"context_lines":[{"line_number":150,"context_line":"Champion"},{"line_number":151,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Sylvain Bauza \u003csbauza@redhat.com\u003e (sbauza)"},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Status \u0026 Tracking"},{"line_number":156,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"aa6ede8b_711ff055","line":153,"range":{"start_line":153,"start_character":0,"end_line":153,"end_character":42},"updated":"2026-05-03 05:50:57.000000000","message":"Honestly I admire your conviction for this. However, I want to suggest a better approach that single handedly signing up to monitor (or modify) all the official OpenStack governed code repositories. \n\nMy approach would be a working group perhaps, or even the TACT-SIG where you\u0027ve gotten the agentic-workflows repo to be added as collaborators.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"28bb939e005e10b9a7f73abd95d7b845ee3a457a","unresolved":true,"context_lines":[{"line_number":150,"context_line":"Champion"},{"line_number":151,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Sylvain Bauza \u003csbauza@redhat.com\u003e (sbauza)"},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Status \u0026 Tracking"},{"line_number":156,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a8e67a3_0f4558a4","line":153,"range":{"start_line":153,"start_character":0,"end_line":153,"end_character":42},"in_reply_to":"aa6ede8b_711ff055","updated":"2026-05-04 20:26:56.000000000","message":"honestly saying, having one member to champian goal is a win here or for any goal :) more than one is better but I do not think community is in that situation.","commit_id":"f4cdf95de437d1c243cbb6b2d6593604df19933e"}]}