How one can Construct and Evolve a Customized OpenAI Agent with A-Evolve Utilizing Benchmarks, Expertise, Reminiscence, and Workspace Mutations

import os
import sys
import json
import textwrap
import subprocess
import shutil
from pathlib import Path
from getpass import getpass
from collections import Counter, defaultdict


subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "openai>=1.30.0", "pyyaml>=6.0", "matplotlib>=3.8"])
REPO_DIR = Path("/content material/a-evolve")
if REPO_DIR.exists():
   shutil.rmtree(REPO_DIR)
subprocess.check_call(["git", "clone", "--depth", "1", "https://github.com/A-EVO-Lab/a-evolve.git", str(REPO_DIR)])
sys.path.insert(0, str(REPO_DIR))


if not os.environ.get("OPENAI_API_KEY"):
   os.environ["OPENAI_API_KEY"] = getpass("Enter your OpenAI API key: ").strip()


OPENAI_MODEL = "gpt-4o-mini"


import yaml
import matplotlib.pyplot as plt


import agent_evolve as ae
from agent_evolve.protocol.base_agent import BaseAgent
from agent_evolve.benchmarks.base import BenchmarkAdapter
from agent_evolve.engine.base import EvolutionEngine
from agent_evolve.sorts import Activity, Trajectory, Suggestions, StepResult
from agent_evolve.contract.workspace import AgentWorkspace
from openai import OpenAI


shopper = OpenAI(api_key=os.environ["OPENAI_API_KEY"])


WORKSPACE_ROOT = Path("/content material/a_evolve_demo_workspace")
if WORKSPACE_ROOT.exists():
   shutil.rmtree(WORKSPACE_ROOT)


(WORKSPACE_ROOT / "prompts").mkdir(dad and mom=True, exist_ok=True)
(WORKSPACE_ROOT / "abilities").mkdir(dad and mom=True, exist_ok=True)
(WORKSPACE_ROOT / "reminiscence").mkdir(dad and mom=True, exist_ok=True)
(WORKSPACE_ROOT / "instruments").mkdir(dad and mom=True, exist_ok=True)


manifest = {
   "identify": "colab-aevolve-demo-agent",
   "model": "0.1.0",
   "contract_version": "1.0",
   "agent": {
       "sort": "customized",
       "entrypoint": None
   },
   "evolvable_layers": ["prompts", "skills", "memory"],
   "reload_strategy": "scorching"
}
with open(WORKSPACE_ROOT / "manifest.yaml", "w") as f:
   yaml.dump(manifest, f, sort_keys=False)


initial_system_prompt = textwrap.dedent("""
You're a exact text-transformation agent.


Clear up the duty precisely.
Be concise.
Return solely the ultimate reply with no clarification until the duty explicitly asks for JSON.
""").strip()


(WORKSPACE_ROOT / "prompts" / "system.md").write_text(initial_system_prompt)

def build_dataset():
   prepare = [
       {
           "id": "train-01",
           "rule": "json_sum",
           "input": "Numbers: 7, 11, 4",
           "answer": '{"sum":22}'
       },
       {
           "id": "train-02",
           "rule": "json_sum",
           "input": "Numbers: 20, 5, 3, 2",
           "answer": '{"sum":30}'
       },
       {
           "id": "train-03",
           "rule": "acronym_upper",
           "input": "Create the acronym from: retrieval augmented generation",
           "answer": "RAG"
       },
       {
           "id": "train-04",
           "rule": "acronym_upper",
           "input": "Create the acronym from: large language model",
           "answer": "LLM"
       },
       cherry"
       ,
       lion,
       {
           "id": "train-07",
           "rule": "vowel_parity",
           "input": "Word: equation",
           "answer": "EVEN"
       },
       {
           "id": "train-08",
           "rule": "vowel_parity",
           "input": "Word: education",
           "answer": "ODD"
       },
   ]


   holdout = [
       {
           "id": "holdout-01",
           "rule": "json_sum",
           "input": "Numbers: 100, 1, 9",
           "answer": '{"sum":110}'
       },
       {
           "id": "holdout-02",
           "rule": "acronym_upper",
           "input": "Create the acronym from: artificial general intelligence",
           "answer": "AGI"
       },
       mango"
       ,
       {
           "id": "holdout-04",
           "rule": "vowel_parity",
           "input": "Word: aeroplane",
           "answer": "ODD"
       },
   ]
   return prepare, holdout


TRAIN_DATA, HOLDOUT_DATA = build_dataset()


def normalize_text(x: str) -> str:
   return x.strip().substitute(" ", "")


class MiniTextBenchmark(BenchmarkAdapter):
   def __init__(self):
       self.prepare = TRAIN_DATA
       self.holdout = HOLDOUT_DATA


   def get_tasks(self, break up: str = "prepare", restrict: int = 10):
       knowledge = self.prepare if break up == "prepare" else self.holdout
       duties = []
       for row in knowledge[:limit]:
           duties.append(
               Activity(
                   id=row["id"],
                   enter=row["input"],
                   metadata={
                       "rule": row["rule"],
                       "reply": row["answer"]
                   }
               )
           )
       return duties


   def consider(self, job: Activity, trajectory: Trajectory):
       pred = trajectory.output.strip()
       gold = job.metadata["answer"].strip()
       success = normalize_text(pred) == normalize_text(gold)
       element = {
           "rule": job.metadata["rule"],
           "gold": gold,
           "pred": pred,
           "enter": job.enter,
           "success": success
       }
       rating = 1.0 if success else 0.0
       return Suggestions(
           success=success,
           rating=rating,
           element=json.dumps(element, ensure_ascii=False),
           uncooked=element
       )


SKILL_ROUTING = {
   "json_sum": ["json", "sum"],
   "acronym_upper": ["acronym", "uppercase"],
   "pipe_unique_sorted_lower": ["unique", "sorted", "lowercase", "pipe"],
   "vowel_parity": ["vowel", "odd", "even", "parity"]
}

class ColabAEResolverAgent(BaseAgent):
   def __init__(self, workspace_dir: str | Path, mannequin: str = OPENAI_MODEL):
       self.mannequin = mannequin
       tremendous().__init__(workspace_dir)


   def _pick_relevant_skills(self, job: Activity):
       rule = job.metadata.get("rule", "")
       chosen = []
       for ability in self.abilities:
           hay = f"{ability.identify} {ability.description}".decrease()
           if rule == "json_sum" and ("json" in hay or "sum" in hay):
               chosen.append(ability)
           elif rule == "acronym_upper" and ("acronym" in hay or "uppercase" in hay):
               chosen.append(ability)
           elif rule == "pipe_unique_sorted_lower" and any(okay in hay for okay in ["unique", "sorted", "lowercase", "pipe"]):
               chosen.append(ability)
           elif rule == "vowel_parity" and any(okay in hay for okay in ["vowel", "odd", "even", "parity"]):
               chosen.append(ability)
       return chosen[:3]


   def remedy(self, job: Activity) -> Trajectory:
       relevant_skills = self._pick_relevant_skills(job)
       relevant_skill_texts = []
       for s in relevant_skills:
           relevant_skill_texts.append(self.get_skill_content(s.identify))


       memory_text = "n".be part of(
           [f"- {m.get('content', '')}" for m in self.memories[-8:]]
       ).strip()


       skill_block = "nn".be part of(relevant_skill_texts).strip()
       if not skill_block:
           skill_block = "(no abilities loaded but)"


       if not memory_text:
           memory_text = "(no reminiscence but)"


       user_prompt = textwrap.dedent(f"""
       TASK RULE: {job.metadata.get("rule")}
       TASK INPUT:
       {job.enter}


       ACTIVE SYSTEM PROMPT:
       {self.system_prompt}


       RELEVANT SKILLS:
       {skill_block}


       RECENT MEMORIES:
       {memory_text}


       Clear up the duty precisely.
       Return solely the ultimate reply.
       """).strip()


       response = shopper.chat.completions.create(
           mannequin=self.mannequin,
           temperature=0,
           messages=[
               {"role": "system", "content": "You are an exact text-transformation agent."},
               {"role": "user", "content": user_prompt}
           ]
       )


       output = (response.selections[0].message.content material or "").strip()


       self.bear in mind(
           content material=f"Activity {job.id} underneath rule {job.metadata.get('rule')} produced output: {output}",
           class="episodic"
       )


       return Trajectory(
           task_id=job.id,
           output=output,
           steps=[
               {
                   "rule": task.metadata.get("rule"),
                   "used_skills": [s.name for s in relevant_skills],
                   "system_prompt_chars": len(self.system_prompt),
                   "memory_items_seen": len(self.recollections)
               }
           ]
       )


SKILL_TEMPLATES = {
   "json_sum": textwrap.dedent("""
       ---
       identify: json-sum-exact
       description: Add all integers and output strict compact JSON with the one key sum.
       ---
       # JSON Sum Precise


       Process:
       1. Extract all integers from the duty enter.
       2. Add them.
       3. Return precisely one compact JSON object on this format:
          {"sum":NUMBER}
       4. Don't add areas, explanations, markdown, or further keys.
   """).strip(),


   "acronym_upper": textwrap.dedent("""
       ---
       identify: acronym-upper-exact
       description: Construct an uppercase acronym by taking the primary letter of every phrase.
       ---
       # Acronym Higher Precise


       Process:
       1. Establish the phrase after the colon.
       2. Take the primary letter of every phrase.
       3. Convert each letter to uppercase.
       4. Return solely the ultimate acronym, with no punctuation or clarification.
   """).strip(),


   "pipe_unique_sorted_lower": textwrap.dedent("""
       ---
       identify: pipe-unique-sorted-lower
       description: Normalize tokens to lowercase, deduplicate them, type ascending, and be part of them with pipes.
       ---
       # Pipe Distinctive Sorted Decrease


       Process:
       1. Learn the token record after the colon.
       2. Break up by commas.
       3. Trim areas and lowercase each token.
       4. Take away duplicates.
       5. Kind alphabetically ascending.
       6. Be part of with "|" and return solely the ultimate string.
   """).strip(),


   "vowel_parity": textwrap.dedent("""
       ---
       identify: vowel-parity-exact
       description: Depend vowels within the phrase and output ODD or EVEN solely.
       ---
       # Vowel Parity Precise


       Process:
       1. Learn the goal phrase after the colon.
       2. Depend vowels utilizing a, e, i, o, u.
       3. If the rely is odd, output ODD.
       4. If the rely is even, output EVEN.
       5. Return solely ODD or EVEN with no further textual content.
   """).strip(),
}


PROMPT_APPENDIX = textwrap.dedent("""
## STRICT OUTPUT CONTRACT
- Output solely the ultimate reply.
- By no means clarify your reasoning.
- If a job expects JSON, return compact JSON with precise keys solely.
- When a related ability exists, observe it actually.
- Precise format is extra vital than being conversational.
""").strip()

class ColabMutationEngine(EvolutionEngine):
   def __init__(self):
       self.cycle_count = 0


   def step(self, workspace: AgentWorkspace, observations, historical past, trial):
       self.cycle_count += 1


       failed_by_rule = defaultdict(record)
       for obs in observations:
           if not obs.suggestions.success:
               failed_by_rule[obs.task.metadata["rule"]].append({
                   "task_id": obs.job.id,
                   "enter": obs.job.enter,
                   "gold": obs.job.metadata["answer"],
                   "pred": obs.trajectory.output
               })


       mutated = False
       summaries = []


       current_prompt = workspace.read_prompt()
       if "STRICT OUTPUT CONTRACT" not in current_prompt:
           workspace.write_prompt(current_prompt.rstrip() + "nn" + PROMPT_APPENDIX + "n")
           mutated = True
           summaries.append("immediate hardened")


       existing_skill_names = {s.identify for s in workspace.list_skills()}


       needed_rule_to_skill_name = {
           "json_sum": "json-sum-exact",
           "acronym_upper": "acronym-upper-exact",
           "pipe_unique_sorted_lower": "pipe-unique-sorted-lower",
           "vowel_parity": "vowel-parity-exact",
       }


       for rule, fails in failed_by_rule.gadgets():
           skill_name = needed_rule_to_skill_name[rule]
           if skill_name not in existing_skill_names:
               workspace.write_skill(skill_name, SKILL_TEMPLATES[rule])
               mutated = True
               summaries.append(f"added ability {skill_name}")


           workspace.add_memory({
               "content material": f"Cycle {self.cycle_count}: rule={rule} failed {len(fails)} time(s). Widespread failure sample: output formatting or process mismatch. Gold examples have to be adopted precisely.",
               "rule": rule,
               "examples": fails[:2]
           }, class="episodic")


       if not failed_by_rule:
           workspace.add_memory({
               "content material": f"Cycle {self.cycle_count}: all present coaching duties succeeded. Protect precise formatting conduct."
           }, class="episodic")


       abstract = " | ".be part of(summaries) if summaries else "no mutation wanted"
       return StepResult(
           mutated=mutated,
           abstract=abstract,
           metadata={
               "failed_rules": record(failed_by_rule.keys()),
               "num_failed_rules": len(failed_by_rule),
               "cycle": self.cycle_count
           }
       )


def evaluate_split(agent, benchmark, break up="prepare"):
   duties = benchmark.get_tasks(break up=break up, restrict=100)
   rows = []
   complete = 0
   appropriate = 0
   for job in duties:
       traj = agent.remedy(job)
       fb = benchmark.consider(job, traj)
       rows.append({
           "task_id": job.id,
           "rule": job.metadata["rule"],
           "enter": job.enter,
           "gold": job.metadata["answer"],
           "pred": traj.output,
           "rating": fb.rating,
           "success": fb.success
       })
       complete += 1
       appropriate += int(fb.success)
   rating = appropriate / max(complete, 1)
   return rating, rows


def print_table(rows, title, max_rows=20):
   print("n" + "=" * 110)
   print(title)
   print("=" * 110)
   proven = rows[:max_rows]
   for r in proven:
       print(f"[{r['task_id']}] rule={r['rule']}")
       print(f"  enter : {r['input']}")
       print(f"  gold  : {r['gold']}")
       print(f"  pred  : {r['pred']}")
       print(f"  rating : {r['score']}  success={r['success']}")
       print("-" * 110)


def show_workspace(root: Path):
   print("n" + "=" * 110)
   print("EVOLVED WORKSPACE SNAPSHOT")
   print("=" * 110)
   for path in sorted(root.rglob("*")):
       rel = path.relative_to(root)
       if path.is_dir():
           print(f"[DIR ] {rel}/")
       else:
           print(f"[FILE] {rel}")


def show_skill_contents(root: Path):
   skill_files = sorted((root / "abilities").glob("*/SKILL.md"))
   print("n" + "=" * 110)
   print("SKILL FILES")
   print("=" * 110)
   if not skill_files:
       print("No ability recordsdata but.")
   for sf in skill_files:
       print(f"n--- {sf.guardian.identify}/SKILL.md ---")
       print(sf.read_text())

benchmark = MiniTextBenchmark()
agent = ColabAEResolverAgent(WORKSPACE_ROOT, mannequin=OPENAI_MODEL)
engine = ColabMutationEngine()


baseline_train_score, baseline_train_rows = evaluate_split(agent, benchmark, break up="prepare")
baseline_holdout_score, baseline_holdout_rows = evaluate_split(agent, benchmark, break up="holdout")


print(f"Baseline prepare rating   : {baseline_train_score:.3f}")
print(f"Baseline holdout rating : {baseline_holdout_score:.3f}")


print_table(baseline_train_rows, "BASELINE TRAIN RESULTS")
print_table(baseline_holdout_rows, "BASELINE HOLDOUT RESULTS")


config = ae.EvolveConfig(
   batch_size=8,
   max_cycles=4,
   egl_window=2
)


evolver = ae.Evolver(
   agent=agent,
   benchmark=benchmark,
   config=config,
   engine=engine
)


consequence = evolver.run(cycles=4)


print("n" + "=" * 110)
print("A-EVOLVE RUN SUMMARY")
print("=" * 110)
print(f"Cycles accomplished : {consequence.cycles_completed}")
print(f"Closing prepare rating: {consequence.final_score:.3f}")
print(f"Rating historical past    : {consequence.score_history}")
print(f"Converged        : {consequence.converged}")


agent.reload_from_fs()
final_train_score, final_train_rows = evaluate_split(agent, benchmark, break up="prepare")
final_holdout_score, final_holdout_rows = evaluate_split(agent, benchmark, break up="holdout")


print(f"nFinal prepare rating   : {final_train_score:.3f}")
print(f"Closing holdout rating : {final_holdout_score:.3f}")


print_table(final_train_rows, "FINAL TRAIN RESULTS")
print_table(final_holdout_rows, "FINAL HOLDOUT RESULTS")


show_workspace(WORKSPACE_ROOT)
show_skill_contents(WORKSPACE_ROOT)


print("n" + "=" * 110)
print("FINAL SYSTEM PROMPT")
print("=" * 110)
print((WORKSPACE_ROOT / "prompts" / "system.md").read_text())


episodic_path = WORKSPACE_ROOT / "reminiscence" / "episodic.jsonl"
if episodic_path.exists():
   print("n" + "=" * 110)
   print("RECENT EPISODIC MEMORY")
   print("=" * 110)
   traces = episodic_path.read_text().strip().splitlines()
   for line in traces[-10:]:
       print(line)


plt.determine(figsize=(8, 4))
plt.plot(vary(1, len(consequence.score_history) + 1), consequence.score_history, marker="o")
plt.xlabel("Evolution cycle")
plt.ylabel("Prepare rating")
plt.title("A-Evolve rating historical past")
plt.grid(True)
plt.present()


print("n" + "=" * 110)
print("COMPARISON")
print("=" * 110)
print(f"Prepare   : {baseline_train_score:.3f} -> {final_train_score:.3f}")
print(f"Holdout : {baseline_holdout_score:.3f} -> {final_holdout_score:.3f}")


improved_rules = []
for earlier than, after in zip(sorted(baseline_train_rows, key=lambda x: x["task_id"]), sorted(final_train_rows, key=lambda x: x["task_id"])):
   if (not earlier than["success"]) and after["success"]:
       improved_rules.append(after["rule"])


print(f"Improved prepare instances by rule: {dict(Counter(improved_rules))}")


print("nDone. This pocket book used the true A-Evolve framework and demonstrated:")
print("1) a sound agent workspace")
print("2) a BaseAgent subclass")
print("3) a BenchmarkAdapter subclass")
print("4) an EvolutionEngine subclass")
print("5) immediate / ability / reminiscence mutations throughout A-Evolve cycles")