- Move TL;DR section before Installation for quick overview
- Wrap 'For LLM Agents' section in collapsible details tag
- Simplify Configuration section (remove verbose schema explanations)
- Move model override example to Features > Agents section
- Add context7 MCP to Korean README

- Remove sisyphusJuniorModel fallback bypass (BUG #1)
  - userModel fallback was incorrectly bypassing category fallbackChain
  - Now correctly uses category requirements before system default

- Fix variant preservation in actualModel fallback path (BUG #2)
  - Variant was lost when categoryModel undefined but actualModel existed
  - Now preserves variant when falling back to actualModel parsing

- Add variant to session continuation (BUG code-yeongyu#3)
  - Session resume didn't pass variant from message history
  - Now extracts resumeVariant and passes as body.variant

- Remove unused inheritedModel parameter (BUG code-yeongyu#4)
  - resolveCategoryConfig had misleading unused parameter
  - Cleaned up function signature and all call sites

- Add variant-model compatibility validation (BUG code-yeongyu#5)
  - No warning when variant might not be supported by provider
  - Added isVariantLikelySupported() validation with console.warn

Test coverage:
- Updated 10+ existing tests to remove inheritedModel references
- Added 9 new tests for variant validation functions
- All 110 tests pass (77 delegate-task + 33 model-requirements)

Task 4 complete - Wave 2:
- Apply normalizeModelID() in parseModelRef (tiers.ts)
- Apply normalizeModelID() in extractModelName (token-estimator.ts)
- Update test expectations for normalized outputs
- Fixes Bug #4: Model name inconsistencies

All normalization tests passing. Backward compatible.

…ate on plan approval

- Change mcb.* to /mcb.* in .gitignore to avoid ignoring source files like src/config/schema/mcb.ts
- Add resumeTeammateWithMessage() call after plan_approved and plan_rejected paths in send-message-tool.ts
- Fixes PR review comments code-yeongyu#3 and code-yeongyu#4 from cubic-dev-ai

…ts, abort before retry

- Swap resolveAgentForSession priority: session agent before event agent (Bug code-yeongyu#2)
- Add pinnedSessionAgentMap so /start-work agent can't be overwritten by SDK events (Bug code-yeongyu#3)
- Abort in-flight request before fallback promptAsync to prevent silent drops (Bug code-yeongyu#1)
- Add session.get fallback for agent resolution when chat.message never fires (Bug code-yeongyu#4)
- Fix agentPattern regex: replace \b with custom boundary for underscore-delimited session IDs
- Revert resolveAgentFromModel reverse-lookup (unreliable with overlapping models)
- Add tests for pinned agents (8), agent-resolver (17), abort ordering, session.get resolution

Adds ecomode as an alternative to ultrawork that prefers Haiku/Sonnet
agents and avoids Opus, targeting pro-plan users who want cost efficiency.

New files:
- skills/ecomode/SKILL.md - Main skill with 15-domain agent routing
- commands/ecomode.md - User docs (aliases: eco, efficient, save-tokens, budget)
- skills/cancel-ecomode/SKILL.md - Cancel skill following existing pattern

Modified:
- docs/CLAUDE.md - Split keyword routing, conflict resolution, default mode config
- skills/omc-setup/SKILL.md - Step 3.7 for default execution mode selection
- skills/cancel/SKILL.md - Ecomode at priority code-yeongyu#4 with ralph-linking

Routing rules:
- "ulw"/"ultrawork" → always ultrawork
- "eco"/"ecomode" → always ecomode
- "fast"/"parallel" → reads defaultExecutionMode from config
- Conflict (both present) → ecomode wins (more restrictive)

Also includes v3.4.0 facelift changes and ultrapilot command documentation.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

…-yeongyu#309)

Critical fixes:
- Bug #1: SubagentStop hook defaulted all agents to 'failed' because SDK
  doesn't provide `success` field. Now defaults to 'completed' when undefined.
- Bug code-yeongyu#4: Token stats lost across TokenTracker instances. Constructor now
  restores session stats from global state for the same session ID.
- Bug code-yeongyu#5: Ultrawork session isolation bypassed when both session IDs were
  undefined (undefined === undefined). Now rejects all falsy session IDs.

High priority fixes:
- Bug code-yeongyu#6: Cancel skill force-clear missed 12+ state files (boulder,
  hud-state, subagent-tracking, checkpoints, etc). Added comprehensive list.
- Bug code-yeongyu#7: HUD semverCompare() returned NaN on pre-release versions like
  "3.9.5-beta". Fixed to use parseInt and handle pre-release ordering.
- Bug code-yeongyu#8: Silent JSON parse failures in critical state readers. Added
  error logging to ralph and ultrawork state readers.
- Bug code-yeongyu#9: Stale task detection had no default behavior when onStaleSession
  callback was not configured. Now auto-cleans after 2x threshold.
- Bug code-yeongyu#10: Hardcoded 3-architect assumption in validation. Extracted to
  REQUIRED_ARCHITECTS constant.

Medium priority fixes:
- Bug code-yeongyu#11: Auto-invoke history used non-atomic writes. Now uses
  atomicWriteJson to prevent corruption from concurrent sessions.
- Bug code-yeongyu#12: Ecomode docs said "all tasks" use Haiku, contradicting the
  escalation paths. Clarified to "most tasks" with upgrade criteria.
- Bug code-yeongyu#13: Added safeUnlinkSync/safeRmSync utilities to prevent ENOENT
  crashes during cleanup operations.
- Bug code-yeongyu#14: State files containing user prompts written with 0644
  permissions. Now writes with 0600 (owner-only read/write).
- Bug code-yeongyu#15: Model names recorded inconsistently (e.g., 'claude-3-5-haiku'
  vs 'claude-haiku-4'). Now normalizes at recording time via exported
  normalizeModelName().

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

…hardening

Fixes from multi-model council audit (7 members, 19 findings, 9 selected):

- Use parseModelString() for cross-provider Anthropic thinking config (#3)
- Update stale AGENTS.md athena directory listing (#4)
- Replace prompt in appendMissingCouncilPrompt instead of appending (#5)
- Extract duplicated session cleanup logic in agent-switch hook (#6)
- Surface skipped council members when >=2 valid members exist (#9)
- Expand fallback handoff regex with negation guards (#11)
- Remove dead council-member agent from agentSources and tests (#12)
- Make runtime council member duplicate check case-insensitive (#14)
- Fix false-positive schema tests by adding required name field (#18)

…tracking

- Remove overly broad /not found/i regex that matched grep 'No matches found' output
- Replace /expect.*to/i test pattern with /AssertionError/i for precise test failure detection
- Narrow trackBuildResult regex from /error/i to /(?:compilation|type)\s+error/i
- Add bash quiet-success handler to reset lastBuildResult from 'fail' to 'unknown'

Fixes P-007 (false-positive error tracking chain) bugs code-yeongyu#1, code-yeongyu#2, code-yeongyu#4

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

The loop detector's check code-yeongyu#4 used consecutiveCount > 1 as a guard, meaning a single retry of one file amidst 22 different reads triggered the breaker (consecutiveToolNameCount=22, consecutiveCount=2). Now requires consecutiveCount >= max(3, ceil(threshold/2)) identical calls — a single retry won't fire but 5+ identical calls still will.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

…hardening

Fixes from multi-model council audit (7 members, 19 findings, 9 selected):

- Use parseModelString() for cross-provider Anthropic thinking config (#3)
- Update stale AGENTS.md athena directory listing (#4)
- Replace prompt in appendMissingCouncilPrompt instead of appending (#5)
- Extract duplicated session cleanup logic in agent-switch hook (#6)
- Surface skipped council members when >=2 valid members exist (#9)
- Expand fallback handoff regex with negation guards (#11)
- Remove dead council-member agent from agentSources and tests (#12)
- Make runtime council member duplicate check case-insensitive (#14)
- Fix false-positive schema tests by adding required name field (#18)

…hardening

Fixes from multi-model council audit (7 members, 19 findings, 9 selected):

- Use parseModelString() for cross-provider Anthropic thinking config (code-yeongyu#3)
- Update stale AGENTS.md athena directory listing (code-yeongyu#4)
- Replace prompt in appendMissingCouncilPrompt instead of appending (code-yeongyu#5)
- Extract duplicated session cleanup logic in agent-switch hook (code-yeongyu#6)
- Surface skipped council members when >=2 valid members exist (code-yeongyu#9)
- Expand fallback handoff regex with negation guards (code-yeongyu#11)
- Remove dead council-member agent from agentSources and tests (code-yeongyu#12)
- Make runtime council member duplicate check case-insensitive (code-yeongyu#14)
- Fix false-positive schema tests by adding required name field (code-yeongyu#18)

Live E2E revealed that OpenCode's tool.execute.after hook receives
input={tool,sessionID,callID} without args. The original args surface
through output.metadata.agent (display name like 'Momus - Plan Critic')
and output.metadata.prompt.

Changes:
- extractSubagentType now inspects args.subagent_type AND
  metadata.agent AND metadata.task.* as fallbacks.
- Added normalizeSubagentName that strips zero-width chars and
  lowercases first token so 'Hephaestus - Deep Agent' matches
  reviewableAgents=['hephaestus'].
- reviewableAgents set is normalized on creation so comparison is
  case-insensitive and whitespace-tolerant.
- implementerSessionID prefers metadata.sessionID/sessionId/session_id
  before regex-scanning the output string.
- extractPrompt now reads metadata.prompt as fallback.

Tests: 9/9 pass (existing mocks pass args, which is still supported
as primary path; metadata path is the fallback used in real runtime).

Validated against live E2E of 'Refactor arquitetural: extrair
validacao' where the pipeline reached delegation code-yeongyu#4 and the agent
successfully dispatched Prometheus -> Momus review -> Hephaestus
implementation. Files created: src/shared/validation.test.js (plan
execution in progress at commit time).

…ing detection

**Justification — Oracle analysis #4:**
Oracle: 'Add constrained critic lenses to Telamon without making it Momus.'
Telamon v1 answered one question: insight or measurement? Telamon v2
inherits gem-critic's analytical lenses while staying purpose-focused.

**What changed — telamon.ts:**
- JSDoc updated with Oracle citation and gem-critic source references
- 5 critical lenses subordinate to the primary insight question:
  1. Assumption audit — hidden assumptions that reduce to measurement
  2. Over-engineering detection — >50% complexity for <10% insight = blocking
  3. Decomposition validity — by analytical step vs by data source
  4. Causal chain presence — cause/effect vs correlation
  5. Scope-to-insight ratio — scope matches promised insight
- Severity framework: blocking | warning | suggestion
  (at least one blocking required for ITERATE)
- what_works acknowledgment — gem-critic pattern: acknowledge strengths
  before listing failures
- Enhanced output format: What Works, Blocking Issues, Warnings sections
  added

**Why this matters:**
Telamon v1 was binary (OKAY/ITERATE with one reason). Telamon v2
produces actionable, tiered feedback that names exactly which dimension
failed and why. The over-engineering rule ('50% complexity for <10%
insight = blocking') is directly from gem-critic and catches the most
common failure mode: plans that add rigor without insight.

**Gem-critic sources:**
- assumption audit (gem-critic.agent.md:51-55)
- challenge dimensions (gem-critic.agent.md:57-70)
- severity synthesis + what_works (gem-critic.agent.md:72-85)
- over-engineering rule (gem-critic.agent.md:148)