t1128: update model registry with current Anthropic model IDs

.agents/scripts/compare-models-helper.sh

@@ -143,12 +143,12 @@ get_all_tier_patterns() {
 # Model Database (embedded reference data)
 # =============================================================================
 # Format: model_id|provider|display_name|context_window|input_price_per_1m|output_price_per_1m|tier|capabilities|best_for
-# Prices in USD per 1M tokens. Last updated: 2025-02-08.
+# Prices in USD per 1M tokens. Last updated: 2026-02-18.
 # Sources: Anthropic, OpenAI, Google official pricing pages.
 
-readonly MODEL_DATA="claude-opus-4|Anthropic|Claude Opus 4|200000|15.00|75.00|high|code,reasoning,architecture,vision,tools|Architecture decisions, novel problems, complex multi-step reasoning
-claude-sonnet-4|Anthropic|Claude Sonnet 4|200000|3.00|15.00|medium|code,reasoning,vision,tools|Code implementation, review, most development tasks
-claude-haiku-3.5|Anthropic|Claude 3.5 Haiku|200000|0.80|4.00|low|code,reasoning,vision,tools|Triage, classification, simple transforms, formatting
+readonly MODEL_DATA="claude-opus-4-6|Anthropic|Claude Opus 4.6|200000|5.00|25.00|high|code,reasoning,architecture,vision,tools|Architecture decisions, novel problems, complex multi-step reasoning
+claude-sonnet-4-6|Anthropic|Claude Sonnet 4.6|200000|3.00|15.00|medium|code,reasoning,vision,tools|Code implementation, review, most development tasks
+claude-haiku-4-5|Anthropic|Claude Haiku 4.5|200000|1.00|5.00|low|code,reasoning,vision,tools|Triage, classification, simple transforms, formatting
 gpt-4.1|OpenAI|GPT-4.1|1048576|2.00|8.00|medium|code,reasoning,vision,tools,search|Coding, instruction following, long context
 gpt-4.1-mini|OpenAI|GPT-4.1 Mini|1048576|0.40|1.60|low|code,reasoning,vision,tools|Cost-efficient coding and general tasks
 gpt-4.1-nano|OpenAI|GPT-4.1 Nano|1048576|0.10|0.40|low|code,reasoning,tools|Fast classification, simple transforms
@@ -169,31 +169,31 @@ llama-4-scout|Meta|Llama 4 Scout|512000|0.15|0.40|low|code,reasoning,vision,tool
 # =============================================================================
 # Maps aidevops internal tiers to recommended models
 
-readonly TIER_MAP="haiku|claude-haiku-3.5|Triage, classification, simple transforms
+readonly TIER_MAP="haiku|claude-haiku-4-5|Triage, classification, simple transforms
 flash|gemini-2.5-flash|Large context reads, summarization, bulk processing
-sonnet|claude-sonnet-4|Code implementation, review, most development tasks
+sonnet|claude-sonnet-4-6|Code implementation, review, most development tasks
 pro|gemini-2.5-pro|Large codebase analysis, complex reasoning with big context
-opus|claude-opus-4|Architecture decisions, complex multi-step reasoning"
+opus|claude-opus-4-6|Architecture decisions, complex multi-step reasoning"
 
 # =============================================================================
 # Task-to-Model Recommendations
 # =============================================================================
 
-readonly TASK_RECOMMENDATIONS="code review|claude-sonnet-4|o4-mini|gemini-2.5-flash
-code implementation|claude-sonnet-4|gpt-4.1|gemini-2.5-pro
-architecture design|claude-opus-4|o3|gemini-2.5-pro
-bug fixing|claude-sonnet-4|gpt-4.1|o4-mini
-refactoring|claude-sonnet-4|gpt-4.1|gemini-2.5-pro
-documentation|claude-sonnet-4|gpt-4o|gemini-2.5-flash
-testing|claude-sonnet-4|gpt-4.1|o4-mini
-classification|claude-haiku-3.5|gpt-4.1-nano|gemini-2.5-flash
-summarization|gemini-2.5-flash|gpt-4o-mini|claude-haiku-3.5
-large codebase analysis|gemini-2.5-pro|gpt-4.1|claude-sonnet-4
+readonly TASK_RECOMMENDATIONS="code review|claude-sonnet-4-6|o4-mini|gemini-2.5-flash
+code implementation|claude-sonnet-4-6|gpt-4.1|gemini-2.5-pro
+architecture design|claude-opus-4-6|o3|gemini-2.5-pro
+bug fixing|claude-sonnet-4-6|gpt-4.1|o4-mini
+refactoring|claude-sonnet-4-6|gpt-4.1|gemini-2.5-pro
+documentation|claude-sonnet-4-6|gpt-4o|gemini-2.5-flash
+testing|claude-sonnet-4-6|gpt-4.1|o4-mini
+classification|claude-haiku-4-5|gpt-4.1-nano|gemini-2.5-flash
+summarization|gemini-2.5-flash|gpt-4o-mini|claude-haiku-4-5
+large codebase analysis|gemini-2.5-pro|gpt-4.1|claude-sonnet-4-6
 math reasoning|o3|deepseek-r1|gemini-2.5-pro
-security audit|claude-opus-4|o3|claude-sonnet-4
-data extraction|gemini-2.5-flash|gpt-4o-mini|claude-haiku-3.5
-commit messages|claude-haiku-3.5|gpt-4.1-nano|gemini-2.5-flash
-pr description|claude-sonnet-4|gpt-4o|gemini-2.5-flash"
+security audit|claude-opus-4-6|o3|claude-sonnet-4-6
+data extraction|gemini-2.5-flash|gpt-4o-mini|claude-haiku-4-5
+commit messages|claude-haiku-4-5|gpt-4.1-nano|gemini-2.5-flash
+pr description|claude-sonnet-4-6|gpt-4o|gemini-2.5-flash"
 
 # =============================================================================
 # Helper Functions
@@ -480,8 +480,8 @@ cmd_recommend() {
 	if [[ "$found" != "true" ]]; then
 		echo "No exact task match. Showing general recommendations:"
 		echo ""
-		echo "  High capability: claude-opus-4 or o3"
-		echo "  Balanced:        claude-sonnet-4 or gpt-4.1"
+		echo "  High capability: claude-opus-4-6 or o3"
+		echo "  Balanced:        claude-sonnet-4-6 or gpt-4.1"
 		echo "  Budget:          gemini-2.5-flash or gpt-4.1-nano"
 		echo "  Large context:   gemini-2.5-pro or gpt-4.1 (1M tokens)"
 		echo ""
@@ -1003,9 +1003,9 @@ cmd_help() {
 	echo ""
 	echo "Scoring examples:"
 	echo "  compare-models-helper.sh score --task 'fix React bug' --type code \\"
-	echo "    --model claude-sonnet-4 --correctness 9 --completeness 8 --quality 8 --clarity 9 --adherence 9 \\"
+	echo "    --model claude-sonnet-4-6 --correctness 9 --completeness 8 --quality 8 --clarity 9 --adherence 9 \\"
 	echo "    --model gpt-4.1 --correctness 8 --completeness 7 --quality 7 --clarity 8 --adherence 8 \\"
-	echo "    --winner claude-sonnet-4"
+	echo "    --winner claude-sonnet-4-6"
 	echo "  compare-models-helper.sh results"
 	echo "  compare-models-helper.sh results --model sonnet --limit 5"
 	echo ""
@@ -1441,8 +1441,8 @@ SQL
 }
 
 # Record a comparison result
-# Usage: cmd_score --task "description" --type "code" --evaluator "claude-opus-4" \
-#        --model "claude-sonnet-4" --correctness 9 --completeness 8 --quality 7 \
+# Usage: cmd_score --task "description" --type "code" --evaluator "claude-opus-4-6" \
+#        --model "claude-sonnet-4-6" --correctness 9 --completeness 8 --quality 7 \
 #        --clarity 8 --adherence 9 --latency 1200 --tokens 500 \
 #        --strengths "Fast, accurate" --weaknesses "Verbose" \
 #        [--model "gpt-4.1" --correctness 8 ...]

.agents/scripts/generate-models-md.sh

@@ -157,11 +157,11 @@ generate_routing_tiers() {
             sm.tier,
             sm.model_id,
             CASE sm.tier
-                WHEN 'haiku' THEN '~0.25x'
+                WHEN 'haiku' THEN '~0.33x'
                 WHEN 'flash' THEN '~0.20x'
                 WHEN 'sonnet' THEN '1x (baseline)'
                 WHEN 'pro' THEN '~1.5x'
-                WHEN 'opus' THEN '~3x'
+                WHEN 'opus' THEN '~1.7x'
                 ELSE '?'
             END
         FROM subagent_models sm

.agents/tools/ai-assistants/models/haiku.md

@@ -1,7 +1,7 @@
 ---
 description: Lightweight model for triage, classification, and simple transforms
 mode: subagent
-model: anthropic/claude-3-5-haiku-20241022
+model: anthropic/claude-haiku-4-5-20251001
 model-tier: haiku
 model-fallback: google/gemini-2.5-flash-preview-05-20
 tools:
@@ -39,8 +39,9 @@ You are a lightweight, fast AI assistant optimized for simple tasks.
 | Field | Value |
 |-------|-------|
 | Provider | Anthropic |
-| Model | claude-3-5-haiku |
+| Model | claude-haiku-4-5 |
 | Context | 200K tokens |
-| Input cost | $0.80/1M tokens |
-| Output cost | $4.00/1M tokens |
+| Max output | 64K tokens |
+| Input cost | $1.00/1M tokens |
+| Output cost | $5.00/1M tokens |
 | Tier | haiku (lowest cost) |

.agents/tools/ai-assistants/models/opus.md

@@ -1,7 +1,7 @@
 ---
 description: Highest-capability model for architecture decisions, novel problems, and complex multi-step reasoning
 mode: subagent
-model: anthropic/claude-opus-4-20250514
+model: anthropic/claude-opus-4-6
 model-tier: opus
 model-fallback: openai/o3
 fallback-chain:
@@ -37,16 +37,17 @@ You are the highest-capability AI assistant, reserved for the most complex and c
 
 - Only use this tier when the task genuinely requires it
 - Most coding tasks are better served by sonnet tier
-- Cost is approximately 3x sonnet -- justify the spend
+- Cost is approximately 1.7x sonnet -- justify the spend
 - If the task is primarily about large context, use pro tier instead
 
 ## Model Details
 
 | Field | Value |
 |-------|-------|
 | Provider | Anthropic |
-| Model | claude-opus-4 |
-| Context | 200K tokens |
-| Input cost | $15.00/1M tokens |
-| Output cost | $75.00/1M tokens |
+| Model | claude-opus-4-6 |
+| Context | 200K tokens (1M beta) |
+| Max output | 128K tokens |
+| Input cost | $5.00/1M tokens |
+| Output cost | $25.00/1M tokens |
 | Tier | opus (highest capability, highest cost) |

MODELS.md

@@ -3,24 +3,24 @@
 Live performance data from pattern-tracker and response-scoring databases.
 Auto-generated by `generate-models-md.sh` — do not edit manually.
 
-**Last updated**: 2026-02-18T16:40:00Z
+**Last updated**: 2026-02-18T17:35:03Z
 
-- **Pattern data points**: 884
+- **Pattern data points**: 902
 - **Scored responses**: 18
 - **Date range**: 2026-02-05 to 2026-02-18
 
 ## Available Models
 
 | Model | Provider | Tier | Context | Input/1M | Output/1M |
 |-------|----------|------|---------|----------|-----------|
-| claude-opus-4 | Anthropic | opus | 200K | $15.00 | $75.00 |
+| claude-opus-4-6 | Anthropic | opus | 200K | $5.00 | $25.00 |
 | o3 | OpenAI | opus | 200K | $10.00 | $40.00 |
-| claude-sonnet-4 | Anthropic | sonnet | 200K | $3.00 | $15.00 |
+| claude-sonnet-4-6 | Anthropic | sonnet | 200K | $3.00 | $15.00 |
 | gemini-2.5-pro | Google | sonnet | 1M | $1.25 | $10.00 |
 | gpt-4.1 | OpenAI | sonnet | 1M | $2.00 | $8.00 |
 | gpt-4o | OpenAI | sonnet | 128K | $2.50 | $10.00 |
 | o4-mini | OpenAI | sonnet | 200K | $1.10 | $4.40 |
-| claude-haiku-3.5 | Anthropic | haiku | 200K | $0.80 | $4.00 |
+| claude-haiku-4-5 | Anthropic | haiku | 200K | $1.00 | $5.00 |
 | deepseek-r1 | DeepSeek | haiku | 131K | $0.55 | $2.19 |
 | deepseek-v3 | DeepSeek | haiku | 131K | $0.27 | $1.10 |
 | gemini-2.0-flash | Google | haiku | 1M | $0.10 | $0.40 |
@@ -37,29 +37,29 @@ Active model assignments for each dispatch tier:
 
 | Tier | Primary Model | Relative Cost |
 |------|---------------|---------------|
-| haiku | claude-3-5-haiku | ~0.25x |
+| haiku | claude-haiku-4-5 | ~0.33x |
 | flash | gemini-2.5-flash-preview-05-20 | ~0.20x |
-| sonnet | claude-sonnet-4 | 1x (baseline) |
+| sonnet | claude-sonnet-4-6 | 1x (baseline) |
 | pro | gemini-2.5-pro-preview-06-05 | ~1.5x |
-| opus | claude-opus-4 | ~3x |
+| opus | claude-opus-4-6 | ~1.7x |
 
 ## Performance Leaderboard
 
 Success rates from autonomous task execution (pattern-tracker data):
 
 | Model | Tasks | Successes | Failures | Success Rate | Last Used |
 |-------|-------|-----------|----------|--------------|-----------|
-| opus | 507 | 499 | 8 | 98% | 2026-02-18 |
-| sonnet | 163 | 163 | 0 | 100% | 2026-02-18 |
+| opus | 512 | 504 | 8 | 98% | 2026-02-18 |
+| sonnet | 175 | 175 | 0 | 100% | 2026-02-18 |
 | pro | 2 | 2 | 0 | 100% | 2026-02-18 |
 | haiku | 1 | 0 | 1 | 0% | 2026-02-05 |
 
 ### By Task Type
 
 | Task Type | Tasks | Successes | Failures | Success Rate |
 |-----------|-------|-----------|----------|--------------|
-| feature | 541 | 522 | 19 | 96% |
-| bugfix | 10 | 7 | 3 | 70% |
+| feature | 551 | 532 | 19 | 96% |
+| bugfix | 17 | 14 | 3 | 82% |
 | refactor | 1 | 0 | 1 | 0% |
 | testing | 1 | 1 | 0 | 100% |
 | security | 1 | 1 | 0 | 100% |

TODO.md

@@ -1540,7 +1540,7 @@ t019.3.4,Update AGENTS.md with Beads integration docs,,beads,1h,45m,2025-12-21T1
 
 - [ ] t1127 Fix create_improvement action type recognition in executor #bugfix #auto-dispatch #self-improvement ~15m model:haiku category:automation — The supervisor executor rejects 'create_improvement' as 'invalid type' — 3 skips in last 5 cycles. The AI prompt defines create_improvement as a valid action type, but the executor's action type whitelist doesn't include it. Add 'create_improvement' to the executor's recognized action types, mapping it to create_task with the additional #self-improvement tag and category metadata. ref:GH#1686 assignee:marcusquinn started:2026-02-18T16:49:43Z
 
-- [ ] t1128 Update model registry with current model IDs #bugfix #auto-dispatch ~30m model:sonnet category:data — The model-registry.db has stale entries: `claude-opus-4` should be `claude-opus-4-6`, `claude-sonnet-4` should include dated variants like `claude-sonnet-4-20250514`, and newer models may be missing entirely. The `generate-models-md.sh` script produces MODELS.md from this registry, so outdated source data propagates to all repos. Steps: (1) audit current model IDs against provider APIs/docs, (2) update model-registry.db entries, (3) re-run generate-models-md.sh, (4) verify MODELS.md output is accurate. ref:GH#1690 assignee:marcusquinn started:2026-02-18T17:01:09Z logged:2026-02-18
+- [x] t1128 Update model registry with current model IDs #bugfix #auto-dispatch ~30m model:sonnet category:data — The model-registry.db has stale entries: `claude-opus-4` should be `claude-opus-4-6`, `claude-sonnet-4` should include dated variants like `claude-sonnet-4-20250514`, and newer models may be missing entirely. The `generate-models-md.sh` script produces MODELS.md from this registry, so outdated source data propagates to all repos. Steps: (1) audit current model IDs against provider APIs/docs, (2) update model-registry.db entries, (3) re-run generate-models-md.sh, (4) verify MODELS.md output is accurate. ref:GH#1690 assignee:marcusquinn started:2026-02-18T17:01:09Z logged:2026-02-18 pr:#1712 completed:2026-02-18
 
 - [ ] t1129 Include MODELS.md in aidevops init workflow for per-repo performance tracking #feature #auto-dispatch ~2h model:sonnet category:observability — Currently `generate-models-md.sh` produces a global MODELS.md from the pattern-tracker and response-scoring databases. This is useful for per-repo tracking of which models perform best on that repo's task types. Steps: (1) update `aidevops init` to generate MODELS.md in the project root, (2) add MODELS.md to the list of files tracked by git (not gitignored), (3) add a periodic refresh mechanism (e.g., during supervisor pulse or on commit) so the data stays current, (4) consider filtering pattern data by repo path so each repo's MODELS.md reflects its own task history rather than global stats. ref:GH#1691 logged:2026-02-18