As we are discussing updating QEMU's AI policy I took the opportunity to grab Paolo's RFC from:
Message-ID: <[email protected]> Date: Thu, 28 May 2026 09:34:12 +0200 Subject: [PATCH] docs/devel: relax policy on AI-generated contributions From: Paolo Bonzini <[email protected]> And make some updates from the initial review. You can also read the original (long) thread: Message-ID: <[email protected]> Subject: on ai generated and code provenance From: "Michael S. Tsirkin" <[email protected]> Date: Sun, 24 May 2026 08:42:39 -0400 for more background on the discussion. The AGENTS.md is now updated to reflect the proposed policy changes although it still refers to the current but likely soon to be changed security policy. The skills have had the "caveman" treatment: https://github.com/juliusbrussee/caveman in an effort to reduce the overall token cost of the skills. This is becoming more relevant as the true cost of inference is starting to be felt. The recent update to Gemini Flash saw a big increase in capabilities but I also saw a big jump in API costs to go with it. I still have the original skills but so far the caveman versions seem just as capable. I've made a few updates to the triage skill but otherwise they are just terser versions of the last set of skills I posted. Previous discussions suggested that maybe the base AGENTS.md and the triage skill could be up-streamed but if we are proposing the relax the policy maybe including the rest of the skills is now useful? Depending on the response we can split off patches for pull requests as needed. Changes since v3: - roll-up policy update patch with review feedback - caveman all the skills - teach triage about versions - various tags Alex. Alex Bennée (11): AGENTS.md: introduce a very basic guide for AI agents AGENTS.md: expand with information on skills, layout and style scripts/expand-macro.py: helper script exploding macros .agents/skills: add qemu-code-explorer skill (caveman) .agents/skills: add qemu-build skill (caveman) .agents/skills: add qemu-testing skill (caveman) .agents/skills: add qemu-code-reviewer skill (caveman) .agents/skills: add qemu-mail-thread skill (caveman) .agents/skills: add qemu-issue-helper skill (caveman) .agents/skills: add qemu-issue-triage agent skill (caveman) MAINTAINERS: add a section for AI agents Paolo Bonzini (1): docs/devel: relax policy on AI-generated contributions MAINTAINERS | 6 + docs/devel/code-provenance.rst | 172 +++++++---- .agents/skills/qemu-build/SKILL.md | 49 ++++ .agents/skills/qemu-code-explorer/SKILL.md | 84 ++++++ .../qemu-code-explorer/evals/evals.json | 26 ++ .agents/skills/qemu-code-reviewer/SKILL.md | 98 +++++++ .agents/skills/qemu-issue-helper/SKILL.md | 47 +++ .agents/skills/qemu-issue-triage/SKILL.md | 100 +++++++ .../qemu-issue-triage/assets/labels.txt | 133 +++++++++ .../skills/qemu-issue-triage/evals/evals.json | 18 ++ .../scripts/update_labels.sh | 16 + .agents/skills/qemu-mail-thread/SKILL.md | 37 +++ .../scripts/qemu_mail_parser.py | 98 +++++++ .agents/skills/qemu-testing/SKILL.md | 70 +++++ .gitignore | 2 + AGENTS.md | 107 +++++++ GEMINI.md | 1 + scripts/expand-macro.py | 274 ++++++++++++++++++ 18 files changed, 1277 insertions(+), 61 deletions(-) create mode 100644 .agents/skills/qemu-build/SKILL.md create mode 100644 .agents/skills/qemu-code-explorer/SKILL.md create mode 100644 .agents/skills/qemu-code-explorer/evals/evals.json create mode 100644 .agents/skills/qemu-code-reviewer/SKILL.md create mode 100644 .agents/skills/qemu-issue-helper/SKILL.md create mode 100644 .agents/skills/qemu-issue-triage/SKILL.md create mode 100644 .agents/skills/qemu-issue-triage/assets/labels.txt create mode 100644 .agents/skills/qemu-issue-triage/evals/evals.json create mode 100755 .agents/skills/qemu-issue-triage/scripts/update_labels.sh create mode 100644 .agents/skills/qemu-mail-thread/SKILL.md create mode 100644 .agents/skills/qemu-mail-thread/scripts/qemu_mail_parser.py create mode 100644 .agents/skills/qemu-testing/SKILL.md create mode 100644 AGENTS.md create mode 120000 GEMINI.md create mode 100755 scripts/expand-macro.py -- 2.47.3
