🚀 (release): Bump to 0.0.1rc5

[Chisanan232]

What changed

Prepares the python-sdk 0.0.1rc5 (PEP 440) / 0.0.1-rc.5 (display) release, mirroring the rc.4 bump. Two atomic commits:

Version literals (🔖 (release): Bump version to 0.0.1rc5)

• pyproject.toml version = "0.0.1rc5"
• agent_assembly/__init__.py __version__ = "0.0.1rc5"
• sonar-project.properties sonar.projectVersion=0.0.1rc5
• uv.lock regenerated via uv lock (single pinned-version line)

Docs version refs (📝 (docs): Bump doc version refs to 0.0.1rc5)

• docs/compatibility/release-notes.md — new ## 0.0.1-rc.5 section: gRPC :50051 register + fail-loud on unregistered (AAASM-4547); relaxed pydantic/protobuf/grpcio floors + CI resolution guard (AAASM-4518); per-framework quick-start docs tabs (AAASM-4558/4559/4560)
• Five framework example pins >=0.0.1rc4 → >=0.0.1rc5 (agno, haystack, llamaindex-tool-policy, microsoft-agent-framework, smolagents)
• docs/guides/container-base-image.md — 4 governed base-image tag examples v0.0.1-rc.4 → v0.0.1-rc.5
• docs/javascripts/source-facts-guard.js — published pre-release range brought current ..0.0.1rc3 → ..0.0.1rc4 (rc.4 is now published; rc.5 is not yet, so the range stops at rc.4). rc.4's docs bump had missed this file.

Deliberately left unchanged: docs/development/docs-macros.md:75 keeps >=0.0.1rc3 — it is an illustrative historical-fact example inside the "What NOT to template" section and must remain a literal.

Why

Keeps the checked-in version literals, lockfile, SonarCloud gate, and docs in lockstep ahead of the release-python.yml workflow_dispatch that will consume core tag v0.0.1-rc.5.

How to verify

Gates run locally on the final tree: uv sync (clean), pytest test/ (792 passed; 1 pre-existing bench-latency flake unrelated to this change), ruff check/ruff format --check on agent_assembly (clean), mypy via the authoritative pre-commit invocation uv run mypy --ignore-missing-imports (no issues). Both commits pass the repo pre-commit hooks.

Reversible prep only — the release publishes via release-python.yml workflow_dispatch (dry-run → real) AFTER the core rc.5 tag exists; DO NOT dispatch yet.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[Chisanan232]

Claude Code review — PR #251 (rc.5 version bump)

Reviewed head 133367b on release/v0.0.1-rc.5-bump. Do not merge/approve — review only.

1. CI — GREEN ✅
All 32 check runs on the head commit are success or skipped (0 failing): CI Success, build-and-test (unit/integration/contract/all), CodeQL, pip-audit, PEP 561, docs build, LangChain contract test, and the coverage/Sonar/Codecov gates (ignored per policy) all pass. No non-acceptance red to fix.

2. Scope — pure version bump, PEP 440 correct ✅
The authoritative PR diff (base 54377b2..head) is exactly 12 files, +36/−14:

• Version literals: pyproject.toml:7, agent_assembly/__init__.py:11, sonar-project.properties:15 all 0.0.1rc5 — PEP 440 form (rc5, no dot) verified correct.
• uv.lock regenerated (single pinned-version line).
• docs/compatibility/release-notes.md — new ## 0.0.1-rc.5 section (AAASM-4547 / 4518 / 4558-4560).
• 5 example pins >=0.0.1rc4 → >=0.0.1rc5 (agno, haystack, llamaindex-tool-policy, microsoft-agent-framework, smolagents).
• docs/guides/container-base-image.md — 4 base-image tags rc.4 → rc.5.
• docs/javascripts/source-facts-guard.js — published-range ..0.0.1rc3 → ..0.0.1rc4 (correct: rc.4 is now published, rc.5 is not yet; comment describes what's actually published).
• docs/development/docs-macros.md:75 correctly left at >=0.0.1rc3 (historical-fact illustrative literal). Confirmed unchanged.

(Note: a local master...HEAD three-dot diff shows extra files — that's stale local master picking up already-merged rc.5 content, not this PR's scope. The GitHub API base diff and the two-commit range agree: 12 files.)

3. Side effects — version/docs only, no runtime change ✅
The only agent_assembly/ change is the __version__ string literal — no behavior change. Local gates on the final tree:

• uv sync — clean (227 resolved, 192 checked).
• pytest test/ — 793 passed, 15 skipped; the test_init_assembly_coldstart_latency bench flake reran and passed (environmental, not a blocker).
• uv run mypy --ignore-missing-imports agent_assembly (authoritative pre-commit invocation) — No issues found.

4. FE — N/A.

Verdict: APPROVE (advisory) ✅

Clean, correctly-scoped release-prep bump; CI green; all local gates pass. Ready to merge whenever the operator is ready — but per the PR body this is reversible prep and the actual publish must wait until the core v0.0.1-rc.5 tag exists, then dispatch release-python.yml. Not merging/approving here per review-only instruction.

Posted by Claude Code (review only — no merge, no approve).

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud