run_agent.py

"""Run the drug-target-validation environment with Qwen as the agent.

This script provides the baseline inference loop used by the dashboard.
It instantiates :class:`DrugTargetEnvironment`, formats each
:class:`ValidationObservation` into a prompt, calls the local Qwen model
to obtain a structured :class:`DrugTargetAction`, and writes the running
state to the dashboard JSON file.
"""

from __future__ import annotations

import json
import os
import re
import time
from pathlib import Path
from typing import Any, Dict, List, Optional

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

from models import (
    ActionType,
    DrugTargetAction,
    ValidationObservation,
    build_agent_observation_context,
    build_agent_system_prompt,
)
from server.hackathon_environment import DrugTargetEnvironment

DASHBOARD_STATE_PATH = Path(__file__).parent / "_dashboard_state.json"
DASHBOARD_CMD_PATH = Path(__file__).parent / "_dashboard_cmd.json"

USE_PIPELINE = os.getenv("RUN_AGENT_USE_PIPELINE", "0").strip().lower() not in {
    "0", "false", "off",
}


def _parse_thinking_flag() -> bool:
    import sys
    if "--no-thinking" in sys.argv:
        return False
    if "--thinking" in sys.argv:
        return True
    return os.getenv("RUN_AGENT_ENABLE_THINKING", "1").strip().lower() not in {
        "0", "false", "off",
    }


ENABLE_THINKING = _parse_thinking_flag()

MODEL_ID = "Qwen/Qwen2.5-3B-Instruct"
MAX_EPISODE_STEPS = int(os.getenv("RUN_AGENT_MAX_EPISODE_STEPS", "20"))
PIPELINE_TASK = "text-generation"

ACTION_TYPES = [a.value for a in ActionType]

ACTION_TYPE_ALIASES: Dict[str, str] = {
    # Expression / omics
    "expression_lookup": ActionType.QUERY_EXPRESSION.value,
    "tissue_expression": ActionType.QUERY_EXPRESSION.value,
    "gtex_query": ActionType.QUERY_EXPRESSION.value,
    "de_analysis": ActionType.DIFFERENTIAL_EXPRESSION.value,
    "differential_analysis": ActionType.DIFFERENTIAL_EXPRESSION.value,
    "pathway": ActionType.PATHWAY_ENRICHMENT.value,
    "coexpression": ActionType.COEXPRESSION_NETWORK.value,
    # Protein
    "structure_lookup": ActionType.PROTEIN_STRUCTURE_LOOKUP.value,
    "alphafold_lookup": ActionType.PROTEIN_STRUCTURE_LOOKUP.value,
    "binding_site": ActionType.BINDING_SITE_ANALYSIS.value,
    "ppi": ActionType.PROTEIN_INTERACTION_NETWORK.value,
    "interaction_network": ActionType.PROTEIN_INTERACTION_NETWORK.value,
    "druggability": ActionType.DRUGGABILITY_SCREEN.value,
    # Clinical
    "trial_lookup": ActionType.CLINICAL_TRIAL_LOOKUP.value,
    "clinical_lookup": ActionType.CLINICAL_TRIAL_LOOKUP.value,
    "toxicity": ActionType.TOXICITY_PANEL.value,
    "tox_panel": ActionType.TOXICITY_PANEL.value,
    "off_target": ActionType.OFF_TARGET_SCREEN.value,
    "selectivity": ActionType.OFF_TARGET_SCREEN.value,
    "stratification": ActionType.PATIENT_STRATIFICATION.value,
    "patient_subset": ActionType.PATIENT_STRATIFICATION.value,
    # Literature
    "literature": ActionType.LITERATURE_SEARCH.value,
    "pubmed": ActionType.LITERATURE_SEARCH.value,
    "synthesis": ActionType.EVIDENCE_SYNTHESIS.value,
    "competitor": ActionType.COMPETITOR_LANDSCAPE.value,
    # Experimental
    "in_vitro": ActionType.IN_VITRO_ASSAY.value,
    "cell_assay": ActionType.IN_VITRO_ASSAY.value,
    "in_vivo": ActionType.IN_VIVO_MODEL.value,
    "mouse_model": ActionType.IN_VIVO_MODEL.value,
    "crispr": ActionType.CRISPR_KNOCKOUT.value,
    "biomarker": ActionType.BIOMARKER_CORRELATION.value,
    # Meta
    "red_flag": ActionType.FLAG_RED_FLAG.value,
    "expert_review": ActionType.REQUEST_EXPERT_REVIEW.value,
    "submit_report": ActionType.SUBMIT_VALIDATION_REPORT.value,
    "final_report": ActionType.SUBMIT_VALIDATION_REPORT.value,
    "submit": ActionType.SUBMIT_VALIDATION_REPORT.value,
}

SYSTEM_PROMPT = build_agent_system_prompt()

# Sensible default investigation order used as a fallback / hint for the
# dashboard prompt rather than a hard pipeline.
STANDARD_PIPELINE_ORDER: List[ActionType] = [
    ActionType.QUERY_EXPRESSION,
    ActionType.DRUGGABILITY_SCREEN,
    ActionType.OFF_TARGET_SCREEN,
    ActionType.TOXICITY_PANEL,
    ActionType.CLINICAL_TRIAL_LOOKUP,
    ActionType.LITERATURE_SEARCH,
    ActionType.SUBMIT_VALIDATION_REPORT,
]

MODEL_RESPONSE_PREVIEW_CHARS = int(
    os.getenv("RUN_AGENT_MODEL_RESPONSE_PREVIEW_CHARS", "240")
)


def compact_preview(value: Any, max_chars: int = 160) -> str:
    try:
        text = json.dumps(value, ensure_ascii=True, sort_keys=True)
    except TypeError:
        text = str(value)
    text = re.sub(r"\s+", " ", text).strip()
    if len(text) <= max_chars:
        return text
    return text[: max_chars - 3] + "..."


def format_observation(obs: ValidationObservation) -> str:
    parts = [
        f"TARGET: {obs.target_gene}  |  INDICATION: {obs.indication}",
        f"DISEASE CONTEXT: {obs.disease_context}",
        f"Step: {obs.step_index}  |  Credits: "
        f"{obs.credits_remaining}/{obs.credits_total}",
    ]
    context = build_agent_observation_context(obs, max_tools=5)
    if context:
        parts.append(context)

    if obs.pipeline_history:
        last5 = obs.pipeline_history[-5:]
        parts.append("Recent history:")
        for h in last5:
            tag = "OK" if h.get("success") else "FAIL"
            line = (
                f"  [{tag}] {h.get('action_type')}: "
                f"{str(h.get('output_summary', ''))[:80]}"
            )
            parts.append(line)
        completed = {
            h.get("action_type") for h in obs.pipeline_history if h.get("success")
        }
        if completed:
            parts.append(
                f"Completed actions (try not to repeat): "
                f"{', '.join(sorted(map(str, completed)))}"
            )
        remaining = [
            a.value for a in STANDARD_PIPELINE_ORDER
            if a.value not in completed
        ]
        if remaining:
            parts.append(f"Suggested next actions (pick one): {', '.join(remaining)}")

    dossier = obs.dossier
    if dossier.flagged_red_flags:
        parts.append(f"Red flags so far: {dossier.flagged_red_flags[:5]}")
    if dossier.expression_findings:
        parts.append(
            f"Expression findings: "
            f"{compact_preview(dossier.expression_findings, 200)}"
        )
    if dossier.protein_findings:
        parts.append(
            f"Protein findings: "
            f"{compact_preview(dossier.protein_findings, 200)}"
        )
    if dossier.safety_findings:
        parts.append(
            f"Safety findings: "
            f"{compact_preview(dossier.safety_findings, 200)}"
        )
    if dossier.clinical_findings:
        parts.append(
            f"Clinical findings: "
            f"{compact_preview(dossier.clinical_findings, 200)}"
        )
    if dossier.experimental_results:
        parts.append(
            f"Experimental results: "
            f"{compact_preview(dossier.experimental_results, 200)}"
        )
    if obs.latest_output and obs.latest_output.data:
        parts.append(
            f"Latest output: {compact_preview(obs.latest_output.data, 200)}"
        )
    if obs.rule_violations:
        parts.append(f"VIOLATIONS: {obs.rule_violations}")

    parts.append(
        'Output ONLY a single JSON object with these exact keys, no '
        'comments, no extra text:\n'
        '{"action_type": "<one of the available actions>", '
        '"parameters": {}, "reasoning": "<why this step now>", '
        '"final_decision": null, "confidence": null}\n'
        'For the final submit_validation_report, set "final_decision" to '
        '"go" or "no_go" and "confidence" to a number in [0, 1].'
    )
    return "\n".join(parts)


# ── JSON parsing helpers (kept compatible with prior small-LLM quirks) ─


def _repair_truncated_json(text: str) -> Optional[str]:
    s = text.strip()
    if not s.startswith("{"):
        return None

    s = re.sub(r',\s*"[^"\n]*$', '', s)
    s = re.sub(r',\s*"[^"\n]*"\s*:\s*$', '', s)

    in_string = False
    escape = False
    for ch in s:
        if escape:
            escape = False
            continue
        if ch == "\\":
            escape = True
            continue
        if ch == '"':
            in_string = not in_string

    if in_string:
        s += '"'

    open_braces = s.count("{") - s.count("}")
    open_brackets = s.count("[") - s.count("]")
    s += "]" * max(0, open_brackets)
    s += "}" * max(0, open_braces)

    try:
        obj = json.loads(s)
        if isinstance(obj, dict):
            return s
    except json.JSONDecodeError:
        pass

    s = re.sub(r',\s*([}\]])', r'\1', s)
    try:
        obj = json.loads(s)
        if isinstance(obj, dict):
            return s
    except json.JSONDecodeError:
        pass
    return None


def _strip_js_comments(text: str) -> str:
    text = re.sub(r'//[^\n]*', '', text)
    text = re.sub(r'/\*.*?\*/', '', text, flags=re.DOTALL)
    return text


def _normalize_jsonish_text(text: str) -> str:
    text = _strip_js_comments(text)
    text = re.sub(r'(?<=:\s)\bNone\b', 'null', text)
    text = re.sub(r'(?<=:\s)\bTrue\b', 'true', text)
    text = re.sub(r'(?<=:\s)\bFalse\b', 'false', text)
    text = re.sub(r'"([^"\n]+?):"\s*,', r'"\1": "",', text)
    return text


def extract_json_object(text: str) -> Optional[Dict[str, Any]]:
    stripped = _normalize_jsonish_text(text).strip()
    if stripped.startswith('"') and stripped.endswith('"'):
        try:
            unwrapped = json.loads(stripped)
        except json.JSONDecodeError:
            unwrapped = None
        if isinstance(unwrapped, str):
            stripped = _normalize_jsonish_text(unwrapped).strip()
    fence_prefix = "```"
    if stripped.startswith(fence_prefix) and stripped.endswith(fence_prefix):
        lines = stripped.splitlines()
        if len(lines) >= 3:
            stripped = "\n".join(lines[1:-1]).strip()

    candidates: List[str] = [stripped]
    start = stripped.find("{")
    while start != -1:
        depth = 0
        for idx in range(start, len(stripped)):
            char = stripped[idx]
            if char == "{":
                depth += 1
            elif char == "}":
                depth -= 1
                if depth == 0:
                    candidates.append(stripped[start:idx + 1])
                    break
        start = stripped.find("{", start + 1)

    repaired = None
    first_brace = stripped.find("{")
    if first_brace != -1:
        repaired = _repair_truncated_json(stripped[first_brace:])
        if repaired is not None:
            candidates.append(repaired)

    candidates.sort(key=len, reverse=True)

    for candidate in candidates:
        try:
            parsed = json.loads(candidate)
        except json.JSONDecodeError:
            continue
        if isinstance(parsed, dict):
            return parsed

    return None


def _edit_distance(a: str, b: str) -> int:
    if len(a) < len(b):
        return _edit_distance(b, a)
    if not b:
        return len(a)
    prev = list(range(len(b) + 1))
    for i, ca in enumerate(a):
        curr = [i + 1]
        for j, cb in enumerate(b):
            curr.append(min(prev[j + 1] + 1, curr[j] + 1, prev[j] + (ca != cb)))
        prev = curr
    return prev[-1]


def get_payload_value(payload: Dict[str, Any], *names: str) -> Any:
    for name in names:
        if name in payload:
            return payload[name]
    lowered = {str(k).lower(): v for k, v in payload.items()}
    for name in names:
        if name.lower() in lowered:
            return lowered[name.lower()]
    for key, value in lowered.items():
        for name in names:
            threshold = max(2, len(name) // 3)
            if _edit_distance(key, name.lower()) <= threshold:
                return value
    return None


def normalize_optional_string(value: Any) -> Optional[str]:
    if value is None or isinstance(value, bool):
        return None
    if isinstance(value, str):
        value = value.strip()
        return value or None
    if isinstance(value, (int, float)):
        return str(value)
    return compact_preview(value, 80)


def normalize_action_type(raw: Any) -> Optional[str]:
    if not isinstance(raw, str):
        return None
    candidate = raw.strip().lower()
    if candidate in ACTION_TYPES:
        return candidate
    if candidate in ACTION_TYPE_ALIASES:
        return ACTION_TYPE_ALIASES[candidate]
    candidate = re.sub(r"[^a-z0-9]+", "_", candidate).strip("_")
    if candidate in ACTION_TYPES:
        return candidate
    if candidate in ACTION_TYPE_ALIASES:
        return ACTION_TYPE_ALIASES[candidate]

    heuristics = [
        (("expression",), ActionType.QUERY_EXPRESSION.value),
        (("differential",), ActionType.DIFFERENTIAL_EXPRESSION.value),
        (("pathway",), ActionType.PATHWAY_ENRICHMENT.value),
        (("coexpression",), ActionType.COEXPRESSION_NETWORK.value),
        (("structure",), ActionType.PROTEIN_STRUCTURE_LOOKUP.value),
        (("binding", "site"), ActionType.BINDING_SITE_ANALYSIS.value),
        (("interaction",), ActionType.PROTEIN_INTERACTION_NETWORK.value),
        (("druggab",), ActionType.DRUGGABILITY_SCREEN.value),
        (("clinical",), ActionType.CLINICAL_TRIAL_LOOKUP.value),
        (("toxic",), ActionType.TOXICITY_PANEL.value),
        (("off",), ActionType.OFF_TARGET_SCREEN.value),
        (("strat",), ActionType.PATIENT_STRATIFICATION.value),
        (("literature",), ActionType.LITERATURE_SEARCH.value),
        (("synthes",), ActionType.EVIDENCE_SYNTHESIS.value),
        (("competitor",), ActionType.COMPETITOR_LANDSCAPE.value),
        (("vitro",), ActionType.IN_VITRO_ASSAY.value),
        (("vivo",), ActionType.IN_VIVO_MODEL.value),
        (("crispr",), ActionType.CRISPR_KNOCKOUT.value),
        (("biomarker",), ActionType.BIOMARKER_CORRELATION.value),
        (("red", "flag"), ActionType.FLAG_RED_FLAG.value),
        (("review",), ActionType.REQUEST_EXPERT_REVIEW.value),
        (("submit",), ActionType.SUBMIT_VALIDATION_REPORT.value),
        (("report",), ActionType.SUBMIT_VALIDATION_REPORT.value),
    ]
    for fragments, normalized in heuristics:
        if all(fragment in candidate for fragment in fragments):
            return normalized
    return None


def parse_action(text: str) -> Optional[DrugTargetAction]:
    d = extract_json_object(text)
    if d is None:
        return None

    action_type = normalize_action_type(get_payload_value(d, "action_type"))
    if action_type is None:
        return None

    parameters = get_payload_value(d, "parameters", "params") or {}
    if not isinstance(parameters, dict):
        parameters = {}

    raw_conf = get_payload_value(d, "confidence")
    confidence: Optional[float]
    if raw_conf is None:
        confidence = None
    else:
        try:
            confidence = float(raw_conf)
            confidence = max(0.0, min(1.0, confidence))
        except (TypeError, ValueError):
            confidence = None

    reasoning = get_payload_value(
        d, "reasoning", "rationale", "justification", "reason"
    )
    if reasoning is not None and not isinstance(reasoning, str):
        reasoning = compact_preview(reasoning, 200)
    reasoning = reasoning or ""

    final_decision = normalize_optional_string(
        get_payload_value(d, "final_decision", "decision", "go_no_go")
    )
    if final_decision is not None:
        final_decision = final_decision.lower().replace("-", "_")
        if final_decision not in {"go", "no_go"}:
            final_decision = None

    return DrugTargetAction(
        action_type=ActionType(action_type),
        parameters=parameters,
        reasoning=reasoning,
        final_decision=final_decision,
        confidence=confidence,
    )


def ensure_terminal_payload(action: DrugTargetAction) -> DrugTargetAction:
    """Make sure a SUBMIT_VALIDATION_REPORT carries a decision + confidence."""
    if action.action_type != ActionType.SUBMIT_VALIDATION_REPORT:
        return action
    decision = action.final_decision or "no_go"
    confidence = action.confidence if action.confidence is not None else 0.5
    return action.model_copy(update={
        "final_decision": decision,
        "confidence": float(max(0.0, min(1.0, confidence))),
    })


# ── Dashboard state writer ────────────────────────────────────────────


def write_dashboard_state(
    env: DrugTargetEnvironment,
    obs: ValidationObservation,
    *,
    step: int,
    cumulative_reward: float,
    model_response: str = "",
    model_thinking: str = "",
    action: Optional[DrugTargetAction] = None,
    gen_time: float = 0.0,
    episode_done: bool = False,
) -> None:
    latent = env._latent
    snapshot: Dict[str, Any] = {
        "timestamp": time.time(),
        "step": step,
        "episode_done": episode_done,
        "cumulative_reward": cumulative_reward,
        "gen_time_s": round(gen_time, 2),
        "model_response_raw": model_response[:600],
        "model_thinking": model_thinking[:800],
        "thinking_enabled": ENABLE_THINKING,
    }

    snapshot["task"] = {
        "target_gene": obs.target_gene,
        "indication": obs.indication,
        "disease_context": obs.disease_context,
        "credits_total": obs.credits_total,
    }

    snapshot["resources"] = {
        "credits_used": obs.credits_total - obs.credits_remaining,
        "credits_remaining": obs.credits_remaining,
        "credits_total": obs.credits_total,
    }

    snapshot["pipeline_history"] = [
        {
            "step_index": h.get("step_index"),
            "action_type": h.get("action_type"),
            "output_summary": str(h.get("output_summary", ""))[:120],
            "success": h.get("success"),
            "quality_score": round(h.get("quality_score", 0.0), 3),
            "credit_cost": h.get("credit_cost", 0),
        }
        for h in obs.pipeline_history
    ]

    if action:
        snapshot["current_action"] = {
            "action_type": action.action_type.value,
            "parameters": action.parameters,
            "reasoning": action.reasoning,
            "final_decision": action.final_decision,
            "confidence": action.confidence,
        }

    if obs.latest_output:
        lo = obs.latest_output
        snapshot["latest_output"] = {
            "summary": lo.summary,
            "success": lo.success,
            "quality_score": round(lo.quality_score, 3),
            "uncertainty": round(lo.uncertainty, 3),
            "warnings": lo.warnings,
            "data_preview": compact_preview(lo.data, 300) if lo.data else None,
        }

    snapshot["dossier"] = obs.dossier.model_dump()
    snapshot["rule_violations"] = obs.rule_violations
    snapshot["reward_breakdown"] = {
        k: round(v, 4) for k, v in obs.step_reward_breakdown.items()
    }

    if latent:
        t = latent.target
        snapshot["latent"] = {
            "target_profile": {
                "expression_level": t.expression_level,
                "tissue_specificity": round(t.tissue_specificity, 3),
                "disease_overexpression": round(t.disease_overexpression, 3),
                "druggability_score": round(t.druggability_score, 3),
                "binding_pocket_quality": t.binding_pocket_quality,
                "has_known_ligands": t.has_known_ligands,
                "allosteric_site_available": t.allosteric_site_available,
                "selectivity_ratio": round(t.selectivity_ratio, 3),
                "off_target_count": t.off_target_count,
                "off_target_genes": t.off_target_genes,
                "toxicity_profile": t.toxicity_profile,
                "toxicity_tissues": t.toxicity_tissues,
                "clinical_precedent": t.clinical_precedent,
                "clinical_stage_reached": t.clinical_stage_reached,
                "competitor_programs": t.competitor_programs,
                "true_viability_score": round(t.true_viability_score, 3),
                "correct_decision": t.correct_decision,
                "key_evidence_dimensions": t.key_evidence_dimensions,
                "misleading_signals": t.misleading_signals,
            },
            "data_quality": latent.data_quality.model_dump(),
            "credits": {
                "credits_used": latent.credits.credits_used,
                "credits_total": latent.credits.credits_total,
                "credits_remaining": latent.credits.credits_remaining,
            },
            "progress": latent.progress.model_dump(),
            "action_call_counts": latent.action_call_counts,
        }

    try:
        DASHBOARD_STATE_PATH.write_text(
            json.dumps(snapshot, indent=2, default=str), encoding="utf-8",
        )
    except Exception:
        pass


def log(msg: str) -> None:
    print(msg, flush=True)


def build_observation_prompt(obs: ValidationObservation) -> str:
    return format_observation(obs)


def run_with_pipeline(pipe, prompt: str) -> str:
    try:
        _pipe_max = 2048 if ENABLE_THINKING else 300
        result = pipe(prompt, max_new_tokens=_pipe_max, return_full_text=False)
    except Exception:
        return ""

    if isinstance(result, list) and result:
        result = result[0]
    if isinstance(result, dict):
        text = result.get("generated_text") or result.get("text") or result.get("answer")
    elif isinstance(result, str):
        text = result
    else:
        text = ""
    return text.strip() if isinstance(text, str) else ""


def resolve_torch_runtime() -> Dict[str, Any]:
    use_cuda = torch.cuda.is_available()
    bf16 = bool(getattr(torch.cuda, "is_bf16_supported", lambda: False)()) if use_cuda else False
    dtype = torch.bfloat16 if bf16 else (
        torch.float16 if use_cuda else torch.float32
    )
    return {
        "use_cuda": use_cuda,
        "device": "cuda:0" if use_cuda else "cpu",
        "dtype": dtype,
        "device_map": "auto" if use_cuda else None,
        "device_name": torch.cuda.get_device_name(0) if use_cuda else "cpu",
    }


def main():
    tokenizer = None
    model = None
    eos_ids: List[int] = []
    active_pipeline = None

    runtime = resolve_torch_runtime()
    log(
        f"Using local model runtime: device={runtime['device']} "
        f"name={runtime['device_name']} dtype={runtime['dtype']}"
    )

    if USE_PIPELINE:
        log(f"Loading pipeline ({PIPELINE_TASK}) for {MODEL_ID} ...")
        try:
            active_pipeline = pipeline(
                PIPELINE_TASK,
                model=MODEL_ID,
                trust_remote_code=True,
                dtype=runtime["dtype"],
                device=0 if runtime["use_cuda"] else -1,
            )
            log("Pipeline loaded.")
        except Exception as exc:
            log(f"Pipeline load failed ({exc}), falling back to tokenizer+model.")

    if active_pipeline is None:
        log(f"Loading tokenizer for {MODEL_ID} ...")
        tokenizer = AutoTokenizer.from_pretrained(
            MODEL_ID, trust_remote_code=True,
        )
        log("Tokenizer loaded. Loading model (this may download files on first run) ...")

        model = AutoModelForCausalLM.from_pretrained(
            MODEL_ID,
            dtype=runtime["dtype"],
            device_map=runtime["device_map"],
            trust_remote_code=True,
        )
        log(f"Model loaded. Device: {model.device}")

        if tokenizer.eos_token_id is not None:
            eos_ids.append(tokenizer.eos_token_id)
        extra = tokenizer.convert_tokens_to_ids(["<|im_end|>", "<|endoftext|>"])
        for tid in extra:
            if isinstance(tid, int) and tid not in eos_ids:
                eos_ids.append(tid)
        log(f"EOS token ids: {eos_ids}")

    def check_dashboard_command() -> Optional[Dict[str, Any]]:
        try:
            raw = DASHBOARD_CMD_PATH.read_text(encoding="utf-8")
            try:
                DASHBOARD_CMD_PATH.unlink(missing_ok=True)
            except OSError:
                pass
            return json.loads(raw)
        except (FileNotFoundError, json.JSONDecodeError):
            return None

    def run_episode(scenario_name: Optional[str] = None):
        env = DrugTargetEnvironment(scenario_name=scenario_name)
        obs = env.reset()

        log("\n" + "=" * 70)
        log(
            f"TARGET: {obs.target_gene} | INDICATION: {obs.indication} | "
            f"Credits: {obs.credits_total}"
        )
        if ENABLE_THINKING:
            log("Reasoning mode: ENABLED")
        log("=" * 70)

        cumulative_reward = 0.0
        write_dashboard_state(env, obs, step=0, cumulative_reward=0.0)

        for step in range(MAX_EPISODE_STEPS):
            cmd = check_dashboard_command()
            if cmd and cmd.get("action") == "restart":
                log("\n[DASHBOARD] Restart requested — ending episode early.")
                break

            user_msg = build_observation_prompt(obs)

            messages = [
                {"role": "system", "content": SYSTEM_PROMPT},
                {"role": "user", "content": user_msg},
            ]

            if active_pipeline is not None:
                prompt = f"{SYSTEM_PROMPT}\n\n{user_msg}"
            else:
                try:
                    prompt = tokenizer.apply_chat_template(
                        messages,
                        tokenize=False,
                        add_generation_prompt=True,
                        enable_thinking=ENABLE_THINKING,
                    )
                except TypeError:
                    prompt = tokenizer.apply_chat_template(
                        messages,
                        tokenize=False,
                        add_generation_prompt=True,
                    )

            t0 = time.time()
            if active_pipeline is not None:
                response = run_with_pipeline(active_pipeline, prompt)
                if not response:
                    response = format_observation(obs)
            else:
                assert tokenizer is not None and model is not None
                inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
                n_input = inputs["input_ids"].shape[1]
                max_new = 2048 if ENABLE_THINKING else 300
                with torch.no_grad():
                    output_ids = model.generate(
                        **inputs,
                        max_new_tokens=max_new,
                        do_sample=True,
                        temperature=0.7,
                        top_p=0.8,
                        top_k=20,
                        repetition_penalty=1.3,
                        eos_token_id=eos_ids if eos_ids else None,
                    )
                new_tokens = output_ids[0][n_input:]
                response = tokenizer.decode(new_tokens, skip_special_tokens=True).strip()
            gen_time = time.time() - t0

            thinking = ""
            if ENABLE_THINKING:
                think_match = re.search(
                    r"<think>(.*?)</think>", response, re.DOTALL
                )
                if think_match:
                    thinking = think_match.group(1).strip()
                    response = response[think_match.end():].strip()
                elif response.startswith("<think>"):
                    parts = response.split("</think>", 1)
                    if len(parts) == 2:
                        thinking = parts[0].replace("<think>", "").strip()
                        response = parts[1].strip()

            is_last_step = (step == MAX_EPISODE_STEPS - 1)

            action = parse_action(response)
            if action is None:
                if is_last_step:
                    log(
                        "\n  [!] Parse failed on final step — forcing "
                        "submit_validation_report."
                    )
                    action = DrugTargetAction(
                        action_type=ActionType.SUBMIT_VALIDATION_REPORT,
                        reasoning="forced terminal report",
                        final_decision="no_go",
                        confidence=0.5,
                    )
                else:
                    log(f"\n  [!] Parse failed, skipping step. Raw: {response[:150]}")
                    continue

            if is_last_step and action.action_type != ActionType.SUBMIT_VALIDATION_REPORT:
                log(
                    f"\n  [!] Final step — overriding {action.action_type.value} "
                    f"with submit_validation_report."
                )
                action = DrugTargetAction(
                    action_type=ActionType.SUBMIT_VALIDATION_REPORT,
                    reasoning="forced terminal report",
                    final_decision="no_go",
                    confidence=action.confidence or 0.5,
                )

            action = ensure_terminal_payload(action)

            log(f"\nStep {step + 1}: {action.action_type.value}  ({gen_time:.1f}s)")
            if thinking:
                log(f"  Thinking: {thinking[:200]}")
            if action.reasoning:
                log(f"  Reasoning: {action.reasoning}")
            else:
                log("  Reasoning: [model did not provide one]")
            if action.parameters:
                log(f"  Parameters: {compact_preview(action.parameters, 200)}")

            obs = env.step(action)

            if obs.latest_output:
                lo = obs.latest_output
                status = "OK" if lo.success else "FAIL"
                log(f"  [{status}] {lo.summary}")
                if lo.warnings:
                    log(f"  Warnings: {lo.warnings}")

            step_reward = obs.reward
            cumulative_reward += step_reward
            log(f"  Reward: {step_reward:+.3f}  (cum: {cumulative_reward:+.3f})")
            log(
                f"  Credits remaining: {obs.credits_remaining}"
                f"/{obs.credits_total}"
            )

            write_dashboard_state(
                env, obs,
                step=step + 1,
                cumulative_reward=cumulative_reward,
                model_response=response,
                model_thinking=thinking,
                action=action,
                gen_time=gen_time,
                episode_done=obs.done,
            )

            if obs.rule_violations:
                log(f"  Violations: {obs.rule_violations}")

            if obs.done:
                break

        log(f"\n{'=' * 70}")
        log("EPISODE COMPLETE" if obs.done else f"MAX STEPS ({MAX_EPISODE_STEPS})")
        log(f"  Steps: {obs.step_index}")
        log(f"  Total reward: {cumulative_reward:+.3f}")
        log(
            f"  Credits used: {obs.credits_total - obs.credits_remaining}"
            f"/{obs.credits_total}"
        )
        log("=" * 70)

    try:
        DASHBOARD_CMD_PATH.unlink(missing_ok=True)
    except OSError:
        pass
    run_episode()

    while True:
        log("\nWaiting for dashboard command (restart / new task) ...")
        while True:
            cmd = check_dashboard_command()
            if cmd:
                break
            time.sleep(1.0)

        action_type = cmd.get("action", "restart")
        if action_type == "quit":
            log("Quit requested.")
            break

        scenario = cmd.get("scenario_name")
        log(f"\n[DASHBOARD] {action_type} — scenario={scenario}")
        run_episode(scenario_name=scenario)


if __name__ == "__main__":
    main()