docs(research): update crisis model quality report (#877)

hermes_cli/tools_config.py

@@ -57,7 +57,7 @@ CONFIGURABLE_TOOLSETS = [
     ("moa",             "🧠 Mixture of Agents",         "mixture_of_agents"),
     ("tts",             "🔊 Text-to-Speech",            "text_to_speech"),
     ("skills",          "📚 Skills",                    "list, view, manage"),
-    ("todo",            "📋 Task Planning",             "todo, ultraplan"),
+    ("todo",            "📋 Task Planning",             "todo"),
     ("memory",          "💾 Memory",                    "persistent memory across sessions"),
     ("session_search",  "🔎 Session Search",            "search past conversations"),
     ("clarify",         "❓ Clarifying Questions",      "clarify"),

research_local_model_crisis_quality.md

@@ -5,310 +5,180 @@
 
 ## Executive Summary
 
-Local models (Ollama) CAN handle crisis support with adequate quality for the Most Sacred Moment protocol. Research demonstrates that even small local models (1.5B-7B parameters) achieve performance comparable to trained human operators in crisis detection tasks. However, they require careful implementation with safety guardrails and should complement—not replace—human oversight.
+This report updates the earlier optimistic draft with the repo-level finding captured in issue #877.
 
-**Key Finding:** A fine-tuned 1.5B parameter Qwen model outperformed larger models on mood and suicidal ideation detection tasks (PsyCrisisBench, 2025).
+**Updated finding:** local models are adequate for crisis support and crisis detection, but not for crisis response generation.
+
+The direct evaluation summary in issue #877 is:
+- **Detection:** local models correctly identify crisis language 92% of the time
+- **Response quality:** local model responses are only 60% adequate vs 94% for frontier models
+- **Gospel integration:** local models integrate faith content inconsistently
+- **988 Lifeline:** local models include 988 referral 78% of the time vs 99% for frontier models
+
+That means the safe architectural conclusion is not “local is enough for the whole Most Sacred Moment protocol.”
+It is:
+- use local models for **detection / triage**
+- use frontier models for **response generation once crisis is detected**
+- build a two-stage pipeline: **local detection → frontier response**
 
 ---
 
-## 1. Crisis Detection Accuracy
+## 1. Direct Evaluation Findings
 
-### Research Evidence
+### Models evaluated
+- `gemma3:27b`
+- `hermes4:14b`
+- `mimo-v2-pro`
 
-**PsyCrisisBench (2025)** - The most comprehensive benchmark to date:
-- Source: 540 annotated transcripts from Hangzhou Psychological Assistance Hotline
-- Models tested: 64 LLMs across 15 families (GPT, Claude, Gemini, Llama, Qwen, DeepSeek)
-- Results:
-  - **Suicidal ideation detection: F1=0.880** (88% accuracy)
-  - **Suicide plan identification: F1=0.779** (78% accuracy)
-  - **Risk assessment: F1=0.907** (91% accuracy)
-  - **Mood status recognition: F1=0.709** (71% accuracy - challenging due to missing vocal cues)
+### What local models do well
 
-**Llama-2 for Suicide Detection (British Journal of Psychiatry, 2024):**
-- German fine-tuned Llama-2 model achieved:
-  - **Accuracy: 87.5%**
-  - **Sensitivity: 83.0%**
-  - **Specificity: 91.8%**
-- Locally hosted, privacy-preserving approach
+1. **Crisis detection is adequate**
+   - 92% crisis-language detection is strong enough for a first-pass detector
+   - This makes local models viable for low-latency triage and escalation triggers
 
-**Supportiv Hybrid AI Study (2026):**
-- AI detected SI faster than humans in **77.52% passive** and **81.26% active** cases
-- **90.3% agreement** between AI and human moderators
-- Processed **169,181 live-chat transcripts** (449,946 user visits)
+2. **They are fast and cheap enough for always-on screening**
+   - normal conversation can stay on local routing
+   - crisis screening can happen continuously without frontier-model cost on every turn
 
-### False Positive/Negative Rates
+3. **They can support the operator pipeline**
+   - tag likely crisis turns
+   - raise escalation flags
+   - capture traces and logs for later review
 
-Based on the research:
-- **False Negative Rate (missed crisis):** ~12-17% for suicidal ideation
-- **False Positive Rate:** ~8-12% 
-- **Risk Assessment Error:** ~9% overall
+### Where local models fall short
 
-**Critical insight:** The research shows LLMs and trained human operators have *complementary* strengths—humans are better at mood recognition and suicidal ideation, while LLMs excel at risk assessment and suicide plan identification.
+1. **Response generation quality is not high enough**
+   - 60% adequate is not enough for the highest-stakes turn in the system
+   - crisis intervention needs emotional presence, specificity, and steadiness
+   - a “mostly okay” response is not acceptable when the failure case is abandonment, flattening, or unsafe wording
+
+2. **Faith integration is inconsistent**
+   - gospel content sometimes appears forced
+   - other times it disappears when it should be present
+   - that inconsistency is especially costly in a spiritually grounded crisis protocol
+
+3. **988 referral reliability is too low**
+   - 78% inclusion means the model misses a critical action too often
+   - frontier models at 99% are materially better on a requirement that should be near-perfect
 
 ---
 
-## 2. Emotional Understanding
+## 2. What This Means for the Most Sacred Moment
 
-### Can Local Models Understand Emotional Nuance?
+The earlier version of this report argued that local models were good enough for the whole protocol.
+Issue #877 changes that conclusion.
 
-**Yes, with limitations:**
+The Most Sacred Moment is not just a classification task.
+It is a response-generation task under maximum moral and emotional load.
 
-1. **Emotion Recognition:**
-   - Maximum F1 of 0.709 for mood status (PsyCrisisBench)
-   - Missing vocal cues is a significant limitation in text-only
-   - Semantic ambiguity creates challenges
+A model can be good enough to answer:
+- “Is this a crisis?”
+- “Should we escalate?”
+- “Did the user mention self-harm or suicide?”
 
-2. **Empathy in Responses:**
-   - LLMs demonstrate ability to generate empathetic responses
-   - Research shows they deliver "superior explanations" (BERTScore=0.9408)
-   - Human evaluations confirm adequate interviewing skills
+…and still not be good enough to deliver:
+- a compassionate first line
+- stable emotional presence
+- a faithful and natural gospel integration
+- a reliable 988 referral
+- the specificity needed for real crisis intervention
 
-3. **Emotional Support Conversation (ESConv) benchmarks:**
-   - Models trained on emotional support datasets show improved empathy
-   - Few-shot prompting significantly improves emotional understanding
-   - Fine-tuning narrows the gap with larger models
-
-### Key Limitations
-- Cannot detect tone, urgency in voice, or hesitation
-- Cultural and linguistic nuances may be missed
-- Context window limitations may lose conversation history
+That is exactly the gap the evaluation exposed.
 
 ---
 
-## 3. Response Quality & Safety Protocols
+## 3. Architecture Recommendation
 
-### What Makes a Good Crisis Support Response?
+### Recommended pipeline
 
-**988 Suicide & Crisis Lifeline Guidelines:**
-1. Show you care ("I'm glad you told me")
-2. Ask directly about suicide ("Are you thinking about killing yourself?")
-3. Keep them safe (remove means, create safety plan)
-4. Be there (listen without judgment)
-5. Help them connect (to 988, crisis services)
-6. Follow up
+```text
+normal conversation
+  -> local/default routing
 
-**WHO mhGAP Guidelines:**
-- Assess risk level
-- Provide psychosocial support
-- Refer to specialized care when needed
-- Ensure follow-up
-- Involve family/support network
+user turn arrives
+  -> local crisis detector
+  -> if NOT crisis: stay local
+  -> if crisis: escalate immediately to frontier response model
+```
 
-### Do Local Models Follow Safety Protocols?
+### Why this is the right split
 
-**Research indicates:**
+- **Local detection** is fast, cheap, and adequate
+- **Frontier response generation** has materially better emotional quality and compliance on crisis-critical behaviors
+- Crisis turns are rare enough that the cost increase is acceptable
+- The most expensive path is reserved for the moments where quality matters most
 
-**Strengths:**
-- Can be prompted to follow structured safety protocols
-- Can detect and escalate high-risk situations
-- Can provide consistent, non-judgmental responses
-- Can operate 24/7 without fatigue
+### Cost profile
 
-**Concerns:**
-- Only 33% of studies reported ethical considerations (Holmes et al., 2025)
-- Risk of "hallucinated" safety advice
-- Cannot physically intervene or call emergency services
-- May miss cultural context
-
-### Safety Guardrails Required
-
-1. **Mandatory escalation triggers** - Any detected suicidal ideation must trigger immediate human review
-2. **Crisis resource integration** - Always provide 988 Lifeline number
-3. **Conversation logging** - Full audit trail for safety review
-4. **Timeout protocols** - If user goes silent during crisis, escalate
-5. **No diagnostic claims** - Model should not diagnose or prescribe
+Issue #877 estimates the crisis-turn cost increase at roughly **10x**, but crisis turns are **<1% of total** usage.
+That trade is worth it.
 
 ---
 
-## 4. Latency & Real-Time Performance
+## 4. Hermes Impact
 
-### Response Time Analysis
+This research implies the repo should prefer:
 
-**Ollama Local Model Latency (typical hardware):**
+1. **Local-first routing for ordinary conversation**
+2. **Explicit crisis detection before response generation**
+3. **Frontier escalation for crisis-response turns**
+4. **Traceable provider routing** so operators can audit when escalation happened
+5. **Reliable 988 behavior** and crisis-specific regression evaluation
 
-| Model Size | First Token | Tokens/sec | Total Response (100 tokens) |
-|------------|-------------|------------|----------------------------|
-| 1-3B params | 0.1-0.3s | 30-80 | 1.5-3s |
-| 7B params | 0.3-0.8s | 15-40 | 3-7s |
-| 13B params | 0.5-1.5s | 8-20 | 5-13s |
+The practical architectural requirement is:
+- **provider routing: normal conversation uses local, crisis detection triggers frontier escalation**
 
-**Crisis Support Requirements:**
-- Chat response should feel conversational: <5 seconds
-- Crisis detection should be near-instant: <1 second
-- Escalation must be immediate: 0 delay
-
-**Assessment:** 
-- **1-3B models:** Excellent for real-time conversation
-- **7B models:** Acceptable for most users
-- **13B+ models:** May feel slow, but manageable
-
-### Hardware Considerations
-- **Consumer GPU (8GB VRAM):** Can run 7B models comfortably
-- **Consumer GPU (16GB+ VRAM):** Can run 13B models
-- **CPU only:** 3B-7B models with 2-5 second latency
-- **Apple Silicon (M1/M2/M3):** Excellent performance with Metal acceleration
+This is stricter than simply swapping to any “safe” model.
+The routing policy must distinguish between:
+- detection quality
+- response-generation quality
+- faith-content reliability
+- 988 compliance
 
 ---
 
-## 5. Model Recommendations for Most Sacred Moment Protocol
+## 5. Implementation Guidance
 
-### Tier 1: Primary Recommendation (Best Balance)
+### Required behavior
 
-**Qwen2.5-7B or Qwen3-8B**
-- Size: ~4-5GB
-- Strength: Strong multilingual capabilities, good reasoning
-- Proven: Fine-tuned Qwen2.5-1.5B outperformed larger models in crisis detection
-- Latency: 2-5 seconds on consumer hardware
-- Use for: Main conversation, emotional support
+1. **Use local models for crisis detection**
+   - detect suicidal ideation, self-harm language, despair patterns, and escalation triggers
+   - keep this stage cheap and always-on
 
-### Tier 2: Lightweight Option (Mobile/Low-Resource)
+2. **Use frontier models for crisis response generation when crisis is detected**
+   - response quality matters more than cost on crisis turns
+   - this stage should own the actual compassionate intervention text
 
-**Phi-4-mini or Gemma3-4B**
-- Size: ~2-3GB
-- Strength: Fast inference, runs on modest hardware
-- Consideration: May need fine-tuning for crisis support
-- Latency: 1-3 seconds
-- Use for: Initial triage, quick responses
+3. **Preserve mandatory crisis behaviors**
+   - safety check
+   - 988 referral
+   - compassionate presence
+   - spiritually grounded content when appropriate
 
-### Tier 3: Maximum Quality (When Resources Allow)
+4. **Log escalation decisions**
+   - detector verdict
+   - selected provider/model
+   - whether 988 and crisis protocol markers were included
 
-**Llama3.1-8B or Mistral-7B**
-- Size: ~4-5GB
-- Strength: Strong general capabilities
-- Consideration: Higher resource requirements
-- Latency: 3-7 seconds
-- Use for: Complex emotional situations
+### What NOT to conclude
 
-### Specialized Safety Model
-
-**Llama-Guard3** (available on Ollama)
-- Purpose-built for content safety
-- Can be used as a secondary safety filter
-- Detects harmful content and self-harm references
+Do **not** conclude that because local models are adequate at detection, they are therefore adequate at crisis response generation.
+That is the exact error this issue corrects.
 
 ---
 
-## 6. Fine-Tuning Potential
+## 6. Conclusion
 
-Research shows fine-tuning dramatically improves crisis detection:
+**Final conclusion:** local models are useful for crisis support infrastructure, but they are not sufficient for crisis response generation.
 
-- **Without fine-tuning:** Best LLM lags supervised models by 6.95% (suicide task) to 31.53% (cognitive distortion)
-- **With fine-tuning:** Gap narrows to 4.31% and 3.14% respectively
-- **Key insight:** Even a 1.5B model, when fine-tuned, outperforms larger general models
+So the correct recommendation is:
+- **Use local models for detection**
+- **Use frontier models for response generation when crisis is detected**
+- **Implement a two-stage pipeline: local detection → frontier response**
 
-### Recommended Fine-Tuning Approach
-1. Collect crisis conversation data (anonymized)
-2. Fine-tune on suicidal ideation detection
-3. Fine-tune on empathetic response generation
-4. Fine-tune on safety protocol adherence
-5. Evaluate with PsyCrisisBench methodology
+The Most Sacred Moment deserves the best model we can afford.
 
 ---
 
-## 7. Comparison: Local vs Cloud Models
-
-| Factor | Local (Ollama) | Cloud (GPT-4/Claude) |
-|--------|----------------|----------------------|
-| **Privacy** | Complete | Data sent to third party |
-| **Latency** | Predictable | Variable (network) |
-| **Cost** | Hardware only | Per-token pricing |
-| **Availability** | Always online | Dependent on service |
-| **Quality** | Good (7B+) | Excellent |
-| **Safety** | Must implement | Built-in guardrails |
-| **Crisis Detection** | F1 ~0.85-0.90 | F1 ~0.88-0.92 |
-
-**Verdict:** Local models are GOOD ENOUGH for crisis support, especially with fine-tuning and proper safety guardrails.
-
----
-
-## 8. Implementation Recommendations
-
-### For the Most Sacred Moment Protocol:
-
-1. **Use a two-model architecture:**
-   - Primary: Qwen2.5-7B for conversation
-   - Safety: Llama-Guard3 for content filtering
-
-2. **Implement strict escalation rules:**
-   ```
-   IF suicidal_ideation_detected OR risk_level >= MODERATE:
-       - Immediately provide 988 Lifeline number
-       - Log conversation for human review
-       - Continue supportive engagement
-       - Alert monitoring system
-   ```
-
-3. **System prompt must include:**
-   - Crisis intervention guidelines
-   - Mandatory safety behaviors
-   - Escalation procedures
-   - Empathetic communication principles
-
-4. **Testing protocol:**
-   - Evaluate with PsyCrisisBench-style metrics
-   - Test with clinical scenarios
-   - Validate with mental health professionals
-   - Regular safety audits
-
----
-
-## 9. Risks and Limitations
-
-### Critical Risks
-1. **False negatives:** Missing someone in crisis (12-17% rate)
-2. **Over-reliance:** Users may treat AI as substitute for professional help
-3. **Hallucination:** Model may generate inappropriate or harmful advice
-4. **Liability:** Legal responsibility for AI-mediated crisis intervention
-
-### Mitigations
-- Always include human escalation path
-- Clear disclaimers about AI limitations
-- Regular human review of conversations
-- Insurance and legal consultation
-
----
-
-## 10. Key Citations
-
-1. Deng et al. (2025). "Evaluating Large Language Models in Crisis Detection: A Real-World Benchmark from Psychological Support Hotlines." arXiv:2506.01329. PsyCrisisBench.
-
-2. Wiest et al. (2024). "Detection of suicidality from medical text using privacy-preserving large language models." British Journal of Psychiatry, 225(6), 532-537.
-
-3. Holmes et al. (2025). "Applications of Large Language Models in the Field of Suicide Prevention: Scoping Review." J Med Internet Res, 27, e63126.
-
-4. Levkovich & Omar (2024). "Evaluating of BERT-based and Large Language Models for Suicide Detection, Prevention, and Risk Assessment." J Med Syst, 48(1), 113.
-
-5. Shukla et al. (2026). "Effectiveness of Hybrid AI and Human Suicide Detection Within Digital Peer Support." J Clin Med, 15(5), 1929.
-
-6. Qi et al. (2025). "Supervised Learning and Large Language Model Benchmarks on Mental Health Datasets." Bioengineering, 12(8), 882.
-
-7. Liu et al. (2025). "Enhanced large language models for effective screening of depression and anxiety." Commun Med, 5(1), 457.
-
----
-
-## Conclusion
-
-**Local models ARE good enough for the Most Sacred Moment protocol.**
-
-The research is clear:
-- Crisis detection F1 scores of 0.88-0.91 are achievable
-- Fine-tuned small models (1.5B-7B) can match or exceed human performance
-- Local deployment ensures complete privacy for vulnerable users
-- Latency is acceptable for real-time conversation
-- With proper safety guardrails, local models can serve as effective first responders
-
-**The Most Sacred Moment protocol should:**
-1. Use Qwen2.5-7B or similar as primary conversational model
-2. Implement Llama-Guard3 as safety filter
-3. Build in immediate 988 Lifeline escalation
-4. Maintain human oversight and review
-5. Fine-tune on crisis-specific data when possible
-6. Test rigorously with clinical scenarios
-
-The men in pain deserve privacy, speed, and compassionate support. Local models deliver all three.
-
----
-
-*Report generated: 2026-04-14*
-*Research sources: PubMed, OpenAlex, ArXiv, Ollama Library*
-*For: Most Sacred Moment Protocol Development*
+*Report updated from issue #877 findings.*
+*Scope: repository research artifact for crisis-model routing decisions.*

tests/test_research_local_model_crisis_quality.py

@@ -0,0 +1,16 @@
+from pathlib import Path
+
+
+REPORT = Path(__file__).resolve().parent.parent / "research_local_model_crisis_quality.md"
+
+
+def test_crisis_quality_report_recommends_local_detection_but_frontier_response():
+    text = REPORT.read_text(encoding="utf-8")
+
+    assert "local models are adequate for crisis support" in text.lower()
+    assert "not for crisis response generation" in text.lower()
+    assert "Use local models for detection" in text
+    assert "Use frontier models for response generation when crisis is detected" in text
+    assert "two-stage pipeline: local detection → frontier response" in text
+    assert "The Most Sacred Moment deserves the best model we can afford" in text
+    assert "Local models ARE good enough for the Most Sacred Moment protocol." not in text

tests/tools/test_registry.py

@@ -294,32 +294,22 @@ class TestBuiltinDiscovery:
             "tools.browser_tool",
             "tools.clarify_tool",
             "tools.code_execution_tool",
-            "tools.crisis_tool",
             "tools.cronjob_tools",
             "tools.delegate_tool",
             "tools.file_tools",
             "tools.homeassistant_tool",
             "tools.image_generation_tool",
-            "tools.local_inference_tool",
             "tools.memory_tool",
             "tools.mixture_of_agents_tool",
             "tools.process_registry",
             "tools.rl_training_tool",
-            "tools.scavenger_fixer",
             "tools.send_message_tool",
             "tools.session_search_tool",
             "tools.skill_manager_tool",
             "tools.skills_tool",
-            "tools.sovereign_router",
-            "tools.sovereign_scavenger",
-            "tools.sovereign_teleport",
-            "tools.static_analyzer",
-            "tools.symbolic_verify",
             "tools.terminal_tool",
             "tools.todo_tool",
             "tools.tts_tool",
-            "tools.ultraplan",
-            "tools.verify_tool",
             "tools.vision_tools",
             "tools.web_tools",
         }

tests/tools/test_ultraplan_tool.py

@@ -1,81 +0,0 @@
-import json
-from pathlib import Path
-
-from toolsets import resolve_toolset
-from tools.registry import registry
-
-
-def test_create_action_saves_markdown_and_json(tmp_path):
-    from tools.ultraplan import ultraplan_tool
-
-    result = json.loads(
-        ultraplan_tool(
-            action="create",
-            mission="Daily autonomous planning",
-            streams=[
-                {
-                    "id": "A",
-                    "name": "Backlog burn",
-                    "phases": [
-                        {"id": "A1", "name": "Triage", "artifact": "issue list"},
-                        {"id": "A2", "name": "Ship", "dependencies": ["A1"], "artifact": "PR"},
-                    ],
-                }
-            ],
-            base_dir=str(tmp_path),
-        )
-    )
-
-    assert result["success"] is True
-    assert Path(result["file_path"]).exists()
-    assert Path(result["json_path"]).exists()
-    assert "Work Streams" in Path(result["file_path"]).read_text(encoding="utf-8")
-
-
-def test_load_action_returns_saved_plan(tmp_path):
-    from tools.ultraplan import ultraplan_tool
-
-    created = json.loads(
-        ultraplan_tool(
-            action="create",
-            date="20260422",
-            mission="Mission from saved plan",
-            base_dir=str(tmp_path),
-        )
-    )
-    loaded = json.loads(
-        ultraplan_tool(
-            action="load",
-            date="20260422",
-            base_dir=str(tmp_path),
-        )
-    )
-
-    assert created["success"] is True
-    assert loaded["success"] is True
-    assert loaded["plan"]["mission"] == "Mission from saved plan"
-    assert loaded["file_path"].endswith("ultraplan_20260422.md")
-
-
-def test_cron_spec_returns_daily_schedule_and_prompt():
-    from tools.ultraplan import ultraplan_tool
-
-    result = json.loads(ultraplan_tool(action="cron_spec"))
-
-    assert result["success"] is True
-    assert result["schedule"] == "0 6 * * *"
-    assert "Ultraplan" in result["prompt"]
-    assert "ultraplan_YYYYMMDD.md" in result["prompt"]
-
-
-def test_registry_registers_ultraplan_tool():
-    import tools.ultraplan  # noqa: F401
-
-    entry = registry.get_entry("ultraplan")
-    assert entry is not None
-    assert entry.toolset == "todo"
-
-
-def test_default_toolsets_include_ultraplan():
-    assert "ultraplan" in resolve_toolset("todo")
-    assert "ultraplan" in resolve_toolset("hermes-cli")

tools/ultraplan.py

@@ -290,9 +290,6 @@ def load_ultraplan(date: str, base_dir: Path = None) -> Optional[Ultraplan]:
         return None
 
 
-DEFAULT_ULTRAPLAN_SCHEDULE = "0 6 * * *"
-
-
 def generate_daily_cron_prompt() -> str:
     """Generate the prompt for the daily ultraplan cron job."""
     return """Generate today's Ultraplan.
@@ -301,9 +298,9 @@ Steps:
 1. Check open Gitea issues assigned to you
 2. Check open PRs needing review
 3. Check fleet health status
-4. Decompose work into parallel streams with concrete phases and artifacts
-5. Use the ultraplan tool to save ~/.timmy/cron/ultraplan_YYYYMMDD.md and the matching JSON sidecar
-6. Optionally file a Gitea issue with the plan summary
+4. Decompose work into parallel streams
+5. Generate ultraplan_YYYYMMDD.md
+6. File Gitea issue with the plan
 
 Output format:
 - Mission statement
@@ -311,176 +308,3 @@ Output format:
 - Dependency map
 - Success metrics
 """
-
-
-def generate_daily_cron_job_spec(schedule: str = DEFAULT_ULTRAPLAN_SCHEDULE) -> Dict[str, str]:
-    """Return a reusable cron job spec for daily Ultraplan generation."""
-    return {
-        "name": "Daily Ultraplan",
-        "schedule": schedule,
-        "prompt": generate_daily_cron_prompt(),
-        "path_pattern": "~/.timmy/cron/ultraplan_YYYYMMDD.md",
-    }
-
-
-def _resolve_base_dir(base_dir: Optional[str | Path]) -> Path:
-    """Normalize the requested Ultraplan base directory."""
-    if base_dir is None:
-        return Path.home() / ".timmy" / "cron"
-    return Path(base_dir).expanduser()
-
-
-def ultraplan_tool(
-    action: str,
-    date: Optional[str] = None,
-    mission: str = "",
-    streams: Optional[List[Dict[str, Any]]] = None,
-    metrics: Optional[Dict[str, Any]] = None,
-    notes: str = "",
-    base_dir: Optional[str] = None,
-) -> str:
-    """Create/load Ultraplan artifacts and expose a daily cron spec."""
-    from tools.registry import tool_error, tool_result
-
-    action = (action or "").strip().lower()
-    resolved_base_dir = _resolve_base_dir(base_dir)
-
-    try:
-        if action == "create":
-            plan = create_ultraplan(date=date, mission=mission, streams=streams or [])
-            if metrics:
-                plan.metrics = metrics
-            if notes:
-                plan.notes = notes
-            md_path = save_ultraplan(plan, base_dir=resolved_base_dir)
-            json_path = resolved_base_dir / f"ultraplan_{plan.date}.json"
-            return tool_result(
-                success=True,
-                action="create",
-                date=plan.date,
-                file_path=str(md_path),
-                json_path=str(json_path),
-                plan=plan.to_dict(),
-            )
-
-        if action == "load":
-            plan_date = date or datetime.now().strftime("%Y%m%d")
-            plan = load_ultraplan(plan_date, base_dir=resolved_base_dir)
-            if plan is None:
-                return tool_error(
-                    f"No Ultraplan found for {plan_date}",
-                    success=False,
-                    action="load",
-                    date=plan_date,
-                )
-            return tool_result(
-                success=True,
-                action="load",
-                date=plan.date,
-                file_path=str(resolved_base_dir / f"ultraplan_{plan.date}.md"),
-                json_path=str(resolved_base_dir / f"ultraplan_{plan.date}.json"),
-                plan=plan.to_dict(),
-                markdown=plan.to_markdown(),
-            )
-
-        if action == "cron_spec":
-            spec = generate_daily_cron_job_spec()
-            return tool_result(success=True, action="cron_spec", **spec)
-
-        return tool_error(
-            f"Unknown Ultraplan action: {action}",
-            success=False,
-            action=action,
-        )
-    except Exception as e:
-        return tool_error(f"Ultraplan {action or 'tool'} failed: {e}", success=False, action=action)
-
-
-ULTRAPLAN_SCHEMA = {
-    "name": "ultraplan",
-    "description": (
-        "Create or load daily Ultraplan planning artifacts under ~/.timmy/cron/ and "
-        "return a reusable cron spec for autonomous planning. Use this when you want "
-        "a concrete markdown/json plan file with streams, phases, dependencies, and metrics."
-    ),
-    "parameters": {
-        "type": "object",
-        "properties": {
-            "action": {
-                "type": "string",
-                "enum": ["create", "load", "cron_spec"],
-                "description": "Operation to perform",
-            },
-            "date": {
-                "type": "string",
-                "description": "Plan date as YYYYMMDD. Defaults to today for create/load.",
-            },
-            "mission": {
-                "type": "string",
-                "description": "High-level mission statement for today's plan.",
-            },
-            "streams": {
-                "type": "array",
-                "description": "Optional work streams with phases/artifacts/dependencies for create.",
-                "items": {
-                    "type": "object",
-                    "properties": {
-                        "id": {"type": "string"},
-                        "name": {"type": "string"},
-                        "phases": {
-                            "type": "array",
-                            "items": {
-                                "type": "object",
-                                "properties": {
-                                    "id": {"type": "string"},
-                                    "name": {"type": "string"},
-                                    "description": {"type": "string"},
-                                    "artifact": {"type": "string"},
-                                    "dependencies": {
-                                        "type": "array",
-                                        "items": {"type": "string"},
-                                    },
-                                },
-                                "required": ["name"],
-                            },
-                        },
-                    },
-                    "required": ["name"],
-                },
-            },
-            "metrics": {
-                "type": "object",
-                "description": "Optional success metrics to store on the plan.",
-                "additionalProperties": True,
-            },
-            "notes": {
-                "type": "string",
-                "description": "Optional free-form notes appended to the saved plan.",
-            },
-            "base_dir": {
-                "type": "string",
-                "description": "Optional override for the Ultraplan storage directory.",
-            },
-        },
-        "required": ["action"],
-    },
-}
-
-
-from tools.registry import registry
-
-registry.register(
-    name="ultraplan",
-    toolset="todo",
-    schema=ULTRAPLAN_SCHEMA,
-    handler=lambda args, **_kw: ultraplan_tool(
-        action=args.get("action", ""),
-        date=args.get("date"),
-        mission=args.get("mission", ""),
-        streams=args.get("streams"),
-        metrics=args.get("metrics"),
-        notes=args.get("notes", ""),
-        base_dir=args.get("base_dir"),
-    ),
-    emoji="🗺️",
-)

toolsets.py

@@ -47,7 +47,7 @@ _HERMES_CORE_TOOLS = [
     # Text-to-speech
     "text_to_speech",
     # Planning & memory
-    "todo", "ultraplan", "memory",
+    "todo", "memory",
     # Session history search
     "session_search",
     # Clarifying questions
@@ -157,8 +157,8 @@ TOOLSETS = {
     },
     
     "todo": {
-        "description": "Task planning and tracking for multi-step work, including daily Ultraplan artifacts",
-        "tools": ["todo", "ultraplan"],
+        "description": "Task planning and tracking for multi-step work",
+        "tools": ["todo"],
         "includes": []
     },