release: record the 2.0.0 publication

[bmdhodl]

Summary

• Adds proof/v2.0.0/PUBLICATION.md: the receipts for the live 2.0.0 release (GitHub release, PyPI files and hashes, attestations, clean install, published-wheel.yml matrix, release email delivery).
• Updates memory/state.md, the roadmap 2.0.0 row, ops/FOLLOWUP.md and inbox/log.md to say 2.0.0 is published.
• Changes the proof/v2.0.0/verify.py state check to look for the published 2.0.0 line.

Related Issues

Follows #849 (release prep) and the v2.0.0 tag.

Proof

• python proof/v2.0.0/verify.py: Verified v2.0.0 release prep (exit 0).
• python scripts/sdk_release_guard.py: pass.
• Preflight, readiness and ruff checks: pass.
• showwork session v200-publication: GREEN, 10/10 claims VERIFIED. Ledger in .showwork/.
• showwork session v200-release-overview: GREEN, 5/5 checks, including a live probe of the release page.
• Publish run 37484497553, published-wheel run 37484673368 and release-content run 37484669205 all passed; details in PUBLICATION.md.

Review Readiness

• Public positioning claims have a source/fact ledger: PUBLICATION.md cites each run, hash and PyPI field.
• N/A: state or lock changes, external API collectors, workflow changes.
• Proof artifacts include command, exit code, platform: see PUBLICATION.md.

Risk And Rollback

Docs and proof only. No SDK, MCP or workflow code changes. Revert the commit to roll back.

Scope

• Related ops/doc(s): ROADMAP, memory/state.md, FOLLOWUP
• Does this change the public API? No
• Does this shift roadmap priority? No. It marks the 2.0.0 row as published.

Checklist

• No SDK code changes, so make check coverage is unchanged
• No new hard dependencies in core SDK
• No hardcoded absolute paths
• N/A: __init__.py exports unchanged

Not done

• No image or video is attached to the GitHub release, on purpose: unsigned assets keep Scorecard Signed-Releases at 0. The overview text went on with Patrick's approval (commit 8e6d5f3).
• OpenAI and Google price rows were checked on 2026-07-15; the publish age gate fails from 2026-10-14, so recheck before the next tag.

🤖 Generated with Claude Code

---

Note

Low Risk
Documentation, proof artifacts, and verifier assertion updates only; no runtime, auth, or publish pipeline code changes.

Overview
This PR records that agentguard47 2.0.0 shipped on 2026-10-06—docs and proof only, no SDK or workflow code.

It adds proof/v2.0.0/PUBLICATION.md as the publication receipt: GitHub release and PyPI metadata, wheel/sdist SHA-256 hashes, Trusted Publishing attestations, clean 3.11 install checks, published-wheel.yml matrix results, release-email delivery, and a Release page section (overview text added with no attached assets).

Ops docs now say 2.0.0 is live: memory/state.md lists 2.0.0 as the published package (1.4.0 for Python 3.9–3.10), the roadmap 2.0.0 row is marked published, ops/FOLLOWUP.md notes publication is done and flags OpenAI/Google price-table recheck before the next tag (publish age gate from 2026-10-14), and inbox/log.md logs PR #849 / v2.0.0.

proof/v2.0.0/verify.py now asserts memory/state.md records the 2.0.0 publication instead of the old “PyPI remains 1.4.0” pre-publish wording.

.showwork/ adds verified claim/session ledgers for the publication session and the release-overview update (including an HTTP probe on the live release page).

^{Reviewed by Cursor Bugbot for commit 8e6d5f3. Bugbot is set up for automated code reviews on this repo. Configure here.}

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 20b725e1-2142-4ad0-9d46-b6754941930d)

[github-actions]

🤖 Claude review

LGTM - no blocking issues.

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: b604995d-1bd6-410c-9bc5-bd8f57539c53)

[github-actions]

🤖 Claude review

LGTM - no blocking issues.