fix: update CEO agent table, clarify Sacred Rule enforcement, add profiler to CLI

[gx-ai-architect]

Summary

Documentation and minor code fixes for CEO prompt and CLI consistency.

Addresses: #796, #798, #800, #801

Changes

1. CEO agent table — Added Failure Analyst and Refiner to the agent role table (were missing despite being spawned in Research/Refine workflows)
2. CEO review criteria — Added assessment criteria for Failure Analyst and Refiner
3. Sacred Rules enforcement note — Documented which rules are programmatically enforced (1-3) vs honor-system (4-9)
4. factory eval clarification — Clarified in permitted actions that factory eval (CLI wrapper) is allowed per Sacred Rule 8
5. CLI argparse — Added profiler to match AgentRole Literal type

Questions for reviewers

@akashgit @gx-ai-architect — need your input on these:

1. Profiler agent: keep or remove?
   profiler has a prompt file (factory/agents/prompts/profiler.md) and is in the AgentRole type, but no workflow invokes it and it wasn't in the CLI argparse. I added it to argparse for consistency — but should we remove the profiler entirely as dead code from the refactoring?

2. Sacred Rules 4-5: implement or accept as honor-system?
   Rules 4 (no test deletion) and 5 (no test quality degradation) have zero programmatic enforcement. I documented this as the current state. Should we:

   • (a) Accept honor-system and keep the doc note
   • (b) File a follow-up to implement guard checks for these rules
   • (c) Downgrade them from "Sacred Rules" to "Guidelines" in the CEO prompt

3. Should factory eval be explicitly in the CEO permitted list?
   The CEO already runs factory eval directly (trace evidence: 3x in bc24771b). I clarified this in the permitted actions. But given Sacred Rule 8's spirit of "delegate ALL technical work" — should eval always go through QA instead?

Test plan

• factory agent --help shows profiler in the role choices
• CEO prompt has 7 rows in agent table (not 5)
• Sacred Rules section has enforcement status note
• No existing tests break

🤖 Generated with Claude Code

[github-actions]

Sentrux Quality Report

Absolute

Scanning ....
[scan] git ls-files: 302 total, 290 kept, 12 dropped (ext:12, meta:0, big:0)
[build_project_map] 290 files, 52 unique dirs, 48 cache misses, 2.5ms
[resolve] 439 resolved, 770 unresolved (of 1209 total specs)
[resolve_imports] project_map 2.6ms, suffix_idx 0.5ms, suffix_resolve 10.3ms, total 13.5ms
[build_graphs] 290 files | maps 1.4ms, imports 13.5ms, calls+inherit 3.4ms, total 18.3ms | 438 import, 4376 call, 0 inherit edges
sentrux check — 2 rules checked

Quality: 4702

✗ [Error] max_cc: 5 function(s) exceed max cyclomatic complexity of 30
    factory/cli.py:cmd_ceo (cc=79)
    factory/study.py:study_project_local (cc=43)
    factory/cli.py:_welcome_wizard (cc=39)
    factory/cli.py:cmd_run (cc=37)
    factory/workflow/validation.py:validate_workflow (cc=31)

✗ 1 violation(s) found

Diff (vs base branch)

Scanning ....
[scan] git ls-files: 302 total, 290 kept, 12 dropped (ext:12, meta:0, big:0)
[build_project_map] 290 files, 52 unique dirs, 48 cache misses, 2.5ms
[resolve] 439 resolved, 770 unresolved (of 1209 total specs)
[resolve_imports] project_map 2.6ms, suffix_idx 0.6ms, suffix_resolve 9.0ms, total 12.2ms
[build_graphs] 290 files | maps 1.3ms, imports 12.3ms, calls+inherit 3.2ms, total 16.7ms | 438 import, 4376 call, 0 inherit edges
sentrux gate — structural regression check

Quality:      4702 -> 4702
Coupling:     0.75 → 0.75
Cycles:       4 → 4
God files:    0 → 0

Distance from Main Sequence: 0.35

✓ No degradation detected

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.12%. Comparing base (0fb5f7d) to head (898ebab).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #803   +/-   ##
=======================================
  Coverage   87.12%   87.12%           
=======================================
  Files          81       81           
  Lines       12209    12209           
=======================================
  Hits        10637    10637           
  Misses       1572     1572           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.