feat(platform): validate IBM Bob in real clients

[tirth8205]

Latest refresh — 18 July 2026

• rebased patch-identically onto main 0a3bd6c; current head 9f07ae4
• all 15 hosted checks passed after this rebase
• Bob Shell and Bob IDE are not installed here; this stays draft and is not mergeable by policy## Status

Draft — do not merge until the real-client checks below pass.

This re-stages the safe IBM Bob portion of PR #540 after it was reverted from main because only automated configuration tests had been run. The original contributor remains credited.

Change

• install the MCP server for Bob Shell at its documented user configuration path
• install the MCP server for Bob IDE at its documented project configuration path
• use command-based STDIO entries
• preserve unrelated configuration during install, reinstall, and uninstall
• document both clients

Automated validation

• IBM Bob install/skills tests: 189 passed
• complete non-daemon suite: 1,913 passed, 1 skipped, 2 expected deselections, 2 xpassed
• both multiprocessing parity checks: 2 passed
• Ruff and git diff --check: clean

Required real-client validation

Do not mark this ready or merge it until all items below are recorded with the released applications:

• Bob Shell discovers the generated MCP server without configuration errors
• Bob Shell invokes a code-review-graph tool successfully
• Bob IDE discovers the project MCP server without configuration errors
• Bob IDE invokes a code-review-graph tool successfully
• reinstall/upgrade preserves unrelated Bob MCP entries in both clients
• uninstall removes only code-review-graph and both clients still load their remaining MCP configuration

Carries the safe intent of PR #540 with mohan-kumar-m2 preserved as co-author.

[github-actions]

code-review-graph review

Overall risk: 0.55 (MEDIUM) — 11 changed function(s)/class(es), 0 affected flow(s), 3 test gap(s)

Risk-scored changes

Risk	Level	Symbol	Location	Tested
0.55	medium	code_review_graph/skills.py::install_platform_configs	code_review_graph/skills.py:433	no
0.45	medium	tests/test_skills.py::TestBobPlatforms.test_all_does_not_create_bob_files_without_detection_signal	tests/test_skills.py:1662	(test)
0.30	low	tests/test_cli_install.py::test_platform_choices_use_current_bob_product_names	tests/test_cli_install.py:101	(test)
0.30	low	tests/test_skills.py::TestBobPlatforms.test_bob_platform_entries_follow_current_ibm_contracts	tests/test_skills.py:1566	(test)
0.30	low	tests/test_skills.py::TestBobPlatforms.find_executable	tests/test_skills.py:1647	no
0.25	low	tests/test_skills.py::TestBobPlatforms	tests/test_skills.py:1563	no
0.25	low	tests/test_skills.py::TestBobPlatforms.test_install_bob_shell_uses_global_settings_without_stdio_type	tests/test_skills.py:1580	(test)
0.25	low	tests/test_skills.py::TestBobPlatforms.test_install_bob_ide_uses_project_settings_without_stdio_type	tests/test_skills.py:1593	(test)
0.25	low	tests/test_skills.py::TestBobPlatforms.test_bob_install_preserves_existing_servers_and_is_idempotent	tests/test_skills.py:1605	(test)
0.25	low	tests/test_skills.py::TestBobPlatforms.test_all_detects_bob_ide_from_workspace_directory	tests/test_skills.py:1628	(test)

Test gaps

• code_review_graph/skills.py::install_platform_configs (code_review_graph/skills.py:433)
• tests/test_skills.py::TestBobPlatforms (tests/test_skills.py:1563)
• tests/test_skills.py::TestBobPlatforms.find_executable (tests/test_skills.py:1647)

Token savings: this graph-backed report used ~88,860 fewer tokens (~97%) than reading every changed file in full (estimated, chars/4 approximation).

---

Powered by code-review-graph — local-first analysis; no code leaves the CI runner.