Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Code-Execution Decision — emit-code-vs-reason · agentsope | DecimalAI

Agent Toolingby agentsope

Code-Execution Decision — emit-code-vs-reason

Decision rubric for when an LM agent should write-and-run code (Program-of-Thought / code interpreter) versus reason in natural language: classify each step as deterministic- computable (emit + execute code, feed the result back) vs judgment (stay in prose).

Never made a case worseMaintained

SkillScore

SkillScore96/100

Test results 96

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

83%96%of test cases passed +13 pts

Without skill19 of 23

With skill22 of 23

3 fixed19 passed either way1 still failing0 broke

Measured on gemini-3.6-flash8/24/2026v1

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskWhat is the primary operational symptom when an agent uses natural language reasoning (under-coding) for complex multi-step math instead of executing code?

Without the skillMissed

With the skillPassed

Same model, same task. Answers are capped and redacted.

Use this skill

Your agent runs on

Adds every file the skill has, pinned so you can update it. No account needed.

pip install "decimalai>=0.15"
decimalai skills add agentsope-agentsop-code-execution-decision
claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"

The same install from Python, for a notebook or a setup script.

Install

pip install "decimalai>=0.15"

Add the skill

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

print("Added. Start Claude Code with the command below.")

Start Claude Code with it

claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"

Type these in a Claude Code session. No account or API key. It writes the skill and any files it bundles, and asks first when its safety rating isn't Passed.

Add the DecimalAI plugin, once

/plugin marketplace add decimal-labs/decimalai-claude-plugins
/plugin install decimalai@decimalai

Install this skill

/decimalai:install agentsope-agentsop-code-execution-decision

The plugin needs python3 and curl on your machine.

No Python or account. Claude Code picks it up in its next session.

Save it into your project

curl -fsSL --create-dirs -o .claude/skills/agentsope-agentsop-code-execution-decision/SKILL.md https://app.decimal.ai/s/agentsope-agentsop-code-execution-decision/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/agentsope-agentsop-code-execution-decision/SKILL.md

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Claude Code docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

import asyncio

from claude_agent_sdk import ClaudeAgentOptions
from decimalai.claude_agent_sdk import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

async def main():
    question = "Hello!"
    async with support.run(question) as run:  # one user message, one run
        async for message in run.query(options=ClaudeAgentOptions(max_turns=6)):
            print(message)

# In a notebook or async code: await main()
asyncio.run(main())

Prefer its own loader? ClaudeAgentOptions(plugins=[skills_plugin()], skills="all") reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

The same install from Python, for a notebook or a setup script. agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

import asyncio

from claude_agent_sdk import ClaudeAgentOptions
from decimalai.claude_agent_sdk import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

async def main():
    question = "Hello!"
    async with support.run(question) as run:  # one user message, one run
        async for message in run.query(options=ClaudeAgentOptions(max_turns=6)):
            print(message)

# In a notebook or async code: await main()
asyncio.run(main())

Prefer its own loader? ClaudeAgentOptions(plugins=[skills_plugin()], skills="all") reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from claude_agent_sdk import ClaudeAgentOptions, query

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
options = ClaudeAgentOptions(system_prompt=f"You are a helpful agent.\n\n{skill}", max_turns=6)

async def main():
    async for message in query(prompt="Hello!", options=options):
        print(message)

# In a notebook or async code: await main()
asyncio.run(main())

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Claude Agent SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai-agents]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

from agents import Agent, Runner
from decimalai.openai_agents import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    name=support.name,
    instructions=support.instructions,  # your prompt, then the skill list
    tools=[*support.tools],             # your tools, then load_skill
)
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

Using SandboxAgent? capabilities=[Shell(), decimalai.openai_agents.sandbox_skills()] points its own loader at the same folder.

Install

pip install "decimalai[openai-agents]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

from agents import Agent, Runner
from decimalai.openai_agents import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    name=support.name,
    instructions=support.instructions,  # your prompt, then the skill list
    tools=[*support.tools],             # your tools, then load_skill
)
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

Using SandboxAgent? capabilities=[Shell(), decimalai.openai_agents.sandbox_skills()] points its own loader at the same folder.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = Agent(name="support", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

OpenAI Agents docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

import json

from openai import OpenAI
from decimalai.openai import agent_config

client = OpenAI()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your tools, as Chat Completions takes them, and what answers each.
TOOLS = [{"type": "function", "function": {
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters": {"type": "object", "properties": {"order_id": {"type": "string"}},
                   "required": ["order_id"]},
}}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    # run.system is your prompt, then the skill list.
    messages = [{"role": "system", "content": run.system}, run.message]
    while True:
        reply = client.chat.completions.create(
            model="gpt-5.4-mini",
            messages=messages,
            tools=[*TOOLS, *run.tools],  # your tools, then load_skill
        ).choices[0].message
        messages.append(reply)
        if not reply.tool_calls:
            break
        for call in reply.tool_calls:
            # load_skill is answered by the run; your own tools by FUNCTIONS.
            function = FUNCTIONS.get(call.function.name)
            messages.append(run.tool_result(call) or {
                "role": "tool", "tool_call_id": call.id,
                "content": function(**json.loads(call.function.arguments)) if function
                else f"There is no tool named {call.function.name!r}.",
            })
print(reply.content)

# The Responses API takes run.response_tools instead of run.tools.

Install

pip install "decimalai[openai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

import json

from openai import OpenAI
from decimalai.openai import agent_config

client = OpenAI()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your tools, as Chat Completions takes them, and what answers each.
TOOLS = [{"type": "function", "function": {
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters": {"type": "object", "properties": {"order_id": {"type": "string"}},
                   "required": ["order_id"]},
}}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    # run.system is your prompt, then the skill list.
    messages = [{"role": "system", "content": run.system}, run.message]
    while True:
        reply = client.chat.completions.create(
            model="gpt-5.4-mini",
            messages=messages,
            tools=[*TOOLS, *run.tools],  # your tools, then load_skill
        ).choices[0].message
        messages.append(reply)
        if not reply.tool_calls:
            break
        for call in reply.tool_calls:
            # load_skill is answered by the run; your own tools by FUNCTIONS.
            function = FUNCTIONS.get(call.function.name)
            messages.append(run.tool_result(call) or {
                "role": "tool", "tool_call_id": call.id,
                "content": function(**json.loads(call.function.arguments)) if function
                else f"There is no tool named {call.function.name!r}.",
            })
print(reply.content)

# The Responses API takes run.response_tools instead of run.tools.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
reply = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[
        {"role": "system", "content": f"You are a helpful agent.\n\n{skill}"},
        {"role": "user", "content": "Hello!"},
    ],
).choices[0].message

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

OpenAI SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[anthropic]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

import anthropic
from decimalai.anthropic import agent_config

client = anthropic.Anthropic()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

question = "Hello!"
with support.run(question) as run:  # one user message, one run
    answer = client.beta.messages.tool_runner(
        model="claude-haiku-5-5",
        max_tokens=4096,
        system=run.system,     # your prompt, then the skill list
        tools=[*run.tools],    # your tools, then load_skill
        messages=[{"role": "user", "content": question}],
    ).until_done()

Install

pip install "decimalai[anthropic]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

import anthropic
from decimalai.anthropic import agent_config

client = anthropic.Anthropic()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

question = "Hello!"
with support.run(question) as run:  # one user message, one run
    answer = client.beta.messages.tool_runner(
        model="claude-haiku-5-5",
        max_tokens=4096,
        system=run.system,     # your prompt, then the skill list
        tools=[*run.tools],    # your tools, then load_skill
        messages=[{"role": "user", "content": question}],
    ).until_done()

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
reply = client.messages.create(
    model="claude-haiku-5-5",
    max_tokens=4096,
    system=f"You are a helpful agent.\n\n{skill}",
    messages=[{"role": "user", "content": "Hello!"}],
)

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Anthropic docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[google-genai]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

from google import genai
from google.genai import types
from decimalai.google_genai import agent_config

client = genai.Client()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your functions, as declarations for the model, and what answers each.
TOOLS = [{"function_declarations": [{
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters_json_schema": {"type": "object", "properties": {"order_id": {"type": "string"}},
                               "required": ["order_id"]},
}]}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    config = types.GenerateContentConfig(
        system_instruction=run.system,     # your prompt, then the skill list
        tools=[*TOOLS, *run.tool_params],  # your functions, then load_skill
    )
    contents = [{"role": "user", "parts": [{"text": run.message}]}]
    while True:
        response = client.models.generate_content(
            model="gemini-3.5-flash", contents=contents, config=config,
        )
        if not response.function_calls:
            break
        contents.append(response.candidates[0].content)
        parts = []
        for call in response.function_calls:
            # load_skill is answered by the run; your own functions by FUNCTIONS.
            # A loop of your own rather than automatic function calling, which
            # raises KeyError when the model asks for a function you did not pass.
            function = FUNCTIONS.get(call.name)
            parts.append(run.tool_result(call) or {"function_response": {
                "name": call.name, "id": call.id,
                "response": {"result": function(**(call.args or {})) if function
                             else f"There is no function named {call.name!r}."},
            }})
        contents.append({"role": "user", "parts": parts})
print(response.text)

Install

pip install "decimalai[google-genai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

from google import genai
from google.genai import types
from decimalai.google_genai import agent_config

client = genai.Client()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your functions, as declarations for the model, and what answers each.
TOOLS = [{"function_declarations": [{
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters_json_schema": {"type": "object", "properties": {"order_id": {"type": "string"}},
                               "required": ["order_id"]},
}]}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    config = types.GenerateContentConfig(
        system_instruction=run.system,     # your prompt, then the skill list
        tools=[*TOOLS, *run.tool_params],  # your functions, then load_skill
    )
    contents = [{"role": "user", "parts": [{"text": run.message}]}]
    while True:
        response = client.models.generate_content(
            model="gemini-3.5-flash", contents=contents, config=config,
        )
        if not response.function_calls:
            break
        contents.append(response.candidates[0].content)
        parts = []
        for call in response.function_calls:
            # load_skill is answered by the run; your own functions by FUNCTIONS.
            # A loop of your own rather than automatic function calling, which
            # raises KeyError when the model asks for a function you did not pass.
            function = FUNCTIONS.get(call.name)
            parts.append(run.tool_result(call) or {"function_response": {
                "name": call.name, "id": call.id,
                "response": {"result": function(**(call.args or {})) if function
                             else f"There is no function named {call.name!r}."},
            }})
        contents.append({"role": "user", "parts": parts})
print(response.text)

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from google import genai
from google.genai import types

client = genai.Client()
skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
response = client.models.generate_content(
    model="gemini-3.5-flash",
    contents="Hello!",
    config=types.GenerateContentConfig(system_instruction=f"You are a helpful agent.\n\n{skill}"),
)

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Gemini API docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[langchain]" langchain-openai
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

from langchain.agents import create_agent
from decimalai.langchain import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = create_agent(
    "openai:gpt-5.4-mini",
    tools=[],                         # your tools; the middleware adds load_skill
    middleware=[support.middleware],  # your prompt, the skill list and load_skill
)
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

Install

pip install "decimalai[langchain]" langchain-openai

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

from langchain.agents import create_agent
from decimalai.langchain import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = create_agent(
    "openai:gpt-5.4-mini",
    tools=[],                         # your tools; the middleware adds load_skill
    middleware=[support.middleware],  # your prompt, the skill list and load_skill
)
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = create_agent("openai:gpt-5.4-mini", tools=[], system_prompt=f"You are a helpful agent.\n\n{skill}")
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

LangChain docs

CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

import decimalai.crewai
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                               # your own CrewAI LLM
    skills=decimalai.crewai.skills(),      # the installed folder
    tools=decimalai.crewai.skill_tools(),  # reads a skill's files, runs its scripts
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

The same install from Python, for a notebook or a setup script. CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

import decimalai.crewai
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                               # your own CrewAI LLM
    skills=decimalai.crewai.skills(),      # the installed folder
    tools=decimalai.crewai.skill_tools(),  # reads a skill's files, runs its scripts
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

No DecimalAI package. CrewAI's own loader reads the folder, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                         # your own CrewAI LLM
    skills=[Path(".agents/skills")],
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

CrewAI docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[adk]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

from decimalai.adk import DecimalAIPlugin, agent_config
from google.adk.agents import LlmAgent
from google.adk.apps import App

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = LlmAgent(
    name="support",          # ADK needs a Python identifier
    model="gemini-3.5-flash",
    instruction=support.instruction,  # your prompt, then the skill list
    tools=[*support.tools],           # your tools, then load_skill
)
# DecimalAIPlugin makes each invocation of runners built from this App one trace.
app = App(name="support", root_agent=agent, plugins=[DecimalAIPlugin()])

Prefer ADK's own loader? SkillToolset(skills=decimalai.adk.skills(), ...) reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

Install

pip install "decimalai[adk]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

from decimalai.adk import DecimalAIPlugin, agent_config
from google.adk.agents import LlmAgent
from google.adk.apps import App

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = LlmAgent(
    name="support",          # ADK needs a Python identifier
    model="gemini-3.5-flash",
    instruction=support.instruction,  # your prompt, then the skill list
    tools=[*support.tools],           # your tools, then load_skill
)
# DecimalAIPlugin makes each invocation of runners built from this App one trace.
app = App(name="support", root_agent=agent, plugins=[DecimalAIPlugin()])

Prefer ADK's own loader? SkillToolset(skills=decimalai.adk.skills(), ...) reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from google.adk.agents import LlmAgent

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = LlmAgent(
    name="support",
    model="gemini-3.5-flash",
    instruction=lambda context: f"You are a helpful agent.\n\n{skill}",
)

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Google ADK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[pydantic-ai]"
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

from pydantic_ai import Agent
from decimalai.pydantic_ai import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    "openai:gpt-4o-mini",
    name=support.name,
    tools=[],                 # your tools; the capability adds load_skill
    capabilities=[support],   # your prompt, the skill list and load_skill
)
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

On the Pydantic AI harness, decimalai.pydantic_ai_harness.skills(project) uses its own loader instead.

Install

pip install "decimalai[pydantic-ai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

from pydantic_ai import Agent
from decimalai.pydantic_ai import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    "openai:gpt-4o-mini",
    name=support.name,
    tools=[],                 # your tools; the capability adds load_skill
    capabilities=[support],   # your prompt, the skill list and load_skill
)
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

On the Pydantic AI harness, decimalai.pydantic_ai_harness.skills(project) uses its own loader instead.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = Agent("openai:gpt-4o-mini", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Pydantic AI docs

Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

import decimalai.deepagents
from deepagents import create_deep_agent

agent = create_deep_agent(model=model, **decimalai.deepagents.skills())
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

The default backend reads files but has no shell. To run a skill's scripts, pass skills(backend=LocalShellBackend(...)).

The same install from Python, for a notebook or a setup script. Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

import decimalai.deepagents
from deepagents import create_deep_agent

agent = create_deep_agent(model=model, **decimalai.deepagents.skills())
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

The default backend reads files but has no shell. To run a skill's scripts, pass skills(backend=LocalShellBackend(...)).

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = create_deep_agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Deep Agents docs

Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

import asyncio

from agent_framework import Agent
from decimalai.agent_framework import skills

agent = Agent(client=client, instructions="You are a helpful agent.", context_providers=[skills()])
asyncio.run(agent.run("Hello!"))

It caches skills by default, so restart the process after decimalai skills update.

The same install from Python, for a notebook or a setup script. Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

import asyncio

from agent_framework import Agent
from decimalai.agent_framework import skills

agent = Agent(client=client, instructions="You are a helpful agent.", context_providers=[skills()])
asyncio.run(agent.run("Hello!"))

It caches skills by default, so restart the process after decimalai skills update.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = Agent(client=client, instructions=f"You are a helpful agent.\n\n{skill}")
asyncio.run(agent.run("Hello!"))

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Microsoft Agent Framework docs

Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents
decimalai skills add agentsope-agentsop-code-execution-decision

Add it to your agent

from decimalai.strands import skills
from strands import Agent
from strands.vended_tools import file_editor, shell

agent = Agent(
    model=model,                # always pass one: the default is Amazon Bedrock
    system_prompt="You are a helpful agent.",
    tools=[file_editor, shell],
    plugins=[skills()],
)
agent("Hello!")

Strands loads an agent's skills once, so build the agent again after decimalai skills update.

The same install from Python, for a notebook or a setup script. Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("agentsope-agentsop-code-execution-decision")
if result.exit_code:
    raise SystemExit(result.problems)

from decimalai.strands import skills
from strands import Agent
from strands.vended_tools import file_editor, shell

agent = Agent(
    model=model,                # always pass one: the default is Amazon Bedrock
    system_prompt="You are a helpful agent.",
    tools=[file_editor, shell],
    plugins=[skills()],
)
agent("Hello!")

Strands loads an agent's skills once, so build the agent again after decimalai skills update.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md").read_text()
agent = Agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent("Hello!")

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Strands docs

Adds every file the skill has, pinned so you can update it. Cursor, Copilot, Cline and Warp read this folder by themselves.

pip install "decimalai>=0.15"
decimalai skills add agentsope-agentsop-code-execution-decision

No Python or account. Cursor, Copilot, Cline and Warp read this folder by themselves.

Save it into your project

curl -fsSL --create-dirs -o .agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md https://app.decimal.ai/s/agentsope-agentsop-code-execution-decision/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/agentsope-agentsop-code-execution-decision/SKILL.md

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Cursor and other editors docs

Fetch the file and give its text to your model as instructions. No account or package needed.

Get the file

curl -fsSL https://app.decimal.ai/s/agentsope-agentsop-code-execution-decision/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/agentsope-agentsop-code-execution-decision.json

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 8/24/2026
Source: agentsope/skillalchemy ↗
License: MIT

Manage it in DecimalAI

Saved prompts and skills, pinned versions, and traces of this skill in your runs. DecimalAI does not run your agent.

Claude Code reads files from your project, so DecimalAI config does not apply there.

On Claude Agent SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI Agents: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Anthropic: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Gemini API: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On LangChain: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On CrewAI: set DECIMAL_API_KEY and add decimalai.init(crewai=True) to record runs and skill use. The skill version stays in your repo.

On Google ADK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Pydantic AI: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Deep Agents: set DECIMAL_API_KEY and add decimalai.init(langchain=True) to record runs and skill use. The skill version stays in your repo.

On Microsoft Agent Framework: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

On Strands: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

Editors read files from your project, so DecimalAI config does not apply there.

Anywhere else: record a use with decimalai.log_skill_activation() inside a trace.

Use this skill

SKILL.md

Code-Execution Decision — emit-code-vs-reason

One-liner: LMs are unreliable calculators but reliable coders. When the answer needs determinism and precision — arithmetic, exact data manipulation, deterministic transforms — emit code and run it. When the answer needs judgment, taste, or open-ended synthesis, reason in natural language. The cost of getting this gate wrong is silent: prose arithmetic hallucinates a plausible-looking wrong number, and over-coding a judgment task burns a sandbox round-trip for nothing.

This is an enhancement overlay. DSPy already gives you dspy.ProgramOfThought (PoT) — the mechanism for write-then-execute. What it does not give you is the decision rubric for when to reach for it. That rubric is this skill. Cross-link the sibling [[agentsop-output-format-by-model]] (which decides how code-shaped content should be serialized) and [[agentsop-test-fix-loop]] (which closes the execute → error → retry loop).

1. 何时激活 (When to activate)

Activate this skill before committing a step to a reasoning strategy whenever the task has a verifiable, deterministic core — or whenever you catch an agent doing arithmetic in prose.

Trigger	Signal
Arithmetic / math	"compute the compound interest", "what's 17.5% of $4,392.18", multi-step word problems, unit conversions, date deltas
Precise data manipulation	"sort these 240 rows by the third column", "dedupe and count", "join these two lists on id", "parse this CSV and sum column B"
Deterministic transforms	regex extraction, string reformatting, base conversion, hashing, sorting, set operations
Symbolic / combinatorial	"how many distinct permutations", "solve this system of equations", calendar/scheduling math
You see a model doing math in prose	"Let me add: 1,204 + 8,991 + ... = 10,195" — almost always worth a code check
Choosing a DSPy module	deciding between `ChainOfThought` and `ProgramOfThought` for a signature [dspy.ai/learn/programming/modules/]

Anti-triggers (do NOT reach for code execution):

The task is judgment: tone, summarization, ranking by quality, "is this reply empathetic?", design tradeoffs, open-ended explanation.
The "computation" is trivial and within the model's reliable range (single-digit arithmetic, a 3-item count) — the sandbox round-trip costs more than it saves.
No sandbox is available and the determinism requirement is soft.
The output's consumer is a human reading prose, and an approximate answer is acceptable.

2. 核心心智模型 (Core mental model)

LMs are unreliable calculators but reliable coders.

A language model predicts the next token, not the correct value. When you ask it to add 48,217 + 9,884 in prose, it emits the most plausible-looking digit sequence — which is frequently wrong, and wrong in a way that looks right. The same model can write 48217 + 9884 as a Python expression flawlessly, because emitting the program is a pattern-matching task it is genuinely good at, and the Python interpreter is a deterministic oracle. This decoupling — model writes the recipe, interpreter computes the result — is the entire thesis of Program-of-Thought (PoT) [arXiv 2211.12588] and PAL [arXiv 2211.10435].

            ┌────────────────────────────────────────────────────────┐
            │  THE GATE: does this answer need determinism/precision?  │
            └────────────────────────────────────────────────────────┘
                       │                                  │
              YES (computable)                     NO (judgment)
                       │                                  │
                       ▼                                  ▼
            ┌─────────────────────┐            ┌─────────────────────┐
            │ EMIT CODE           │            │ REASON IN PROSE     │
            │ model writes recipe │            │ model is the engine │
            │ interpreter = oracle│            │ no oracle exists    │
            └─────────────────────┘            └─────────────────────┘
                       │
                       ▼
            sandbox → run → feed result back into LM → LM narrates/uses it

Two failure modes the gate prevents:

Failure	Mechanism	Symptom
Under-coding (reason when you should compute)	LM hallucinates a calculation it cannot reliably perform	A confident, wrong, plausible-looking number; off-by-one counts; arithmetic that "looks" right
Over-coding (compute when you should reason)	LM wraps a judgment task in a sandbox round-trip that adds no determinism	Wasted latency + cost; brittle code that encodes a subjective rubric as if it were a formula; `print("the tone is friendly")`

Why the asymmetry matters. Under-coding fails silently — the wrong number propagates downstream and no exception fires. Over-coding fails loudly and cheaply — you notice the useless sandbox call. So the default lean, when genuinely uncertain and a verifiable core exists, is toward code. But "genuinely uncertain" is the operative phrase: most judgment tasks are not close calls.

The format corollary (from [[agentsop-output-format-by-model]]): once you decide to emit code, emit it as code in a fenced block / single string field — never nested inside JSON sub-structure. Code-in-JSON measurably degrades the code itself (Aider: 61%→20% on GPT-4 Turbo). The execution decision and the serialization decision are two separate gates; pass both.

3. SOP 工作流 (Decision workflow)

A three-step gate. Run it per step, not per task — one task can have computable steps and judgment steps interleaved.

Step 1 — Classify the step: deterministic-computable vs judgment

Ask: "Is there a single correct answer that a program could verify?"

Yes → computable. Arithmetic, sorting, parsing, counting, set/regex ops, symbolic math, anything with a checkable ground truth. → Step 2.
No → judgment. Quality ranking, tone, summarization, design choice, open synthesis. → reason in prose. Stop here.
Mixed → decompose. "Summarize these sales and tell me the total" = a judgment summary step + a computable total step. Route each independently. The total goes to code; the summary stays prose.

A useful tiebreaker for the "trivial computable" gray zone: if the model would be embarrassed to get it wrong and you'd reach for a calculator yourself, emit code. If you'd do it in your head without a second thought, prose is fine.

Step 2 — If computable, emit + run code

Emit the computation as a real program (Python is the PoT default; the interpreter is the oracle).
Keep the code minimal and side-effect-free for the computation — read inputs, compute, print/return the result. No network, no filesystem unless the task is I/O.
Serialize the code per [[agentsop-output-format-by-model]]: fenced block or single string field, not JSON-nested.

Step 3 — Sandbox → run → feed result back into the LM

Sandbox choice (see §4 op #3, cross-link [[agentsop-output-format-by-model]] for the wire shape): DSPy ProgramOfThought uses a Python interpreter (Deno/PythonInterpreter sandbox in recent versions); OpenAI Code Interpreter runs in a managed container; Anthropic code-execution tool runs in a sandboxed VM; LangChain PythonREPLTool runs in-process and is unsandboxed — treat as untrusted-input-hostile.
Run the code in the sandbox; capture stdout / return value / traceback.
Feed the result back into the LM — this is the step under-coders forget. The code produces a value; the LM must consume it to narrate, format, or chain into the next step. PoT is "code computes, LM contextualizes," not "code replaces the LM."
Error → retry (cross-link [[agentsop-test-fix-loop]]): on traceback, feed the error back to the LM, regenerate the code, re-run. Bound the retries (DSPy ProgramOfThought defaults to max_iters ≈ 3). After the bound, fall back to prose reasoning or surface the failure — do not loop forever.

Exit criterion: the step produces either (a) a code-derived value the LM has consumed, or (b) a prose judgment, with the gate decision recorded so a reviewer can audit why code was or wasn't used.

4. 操作模型 (Operations: Trigger → Action → Output → Evidence)

Seven operations. Each row is a reusable move.

#	Op	Trigger	Action	Output	Evidence
1	Computable-vs-judgment gate	Any step about to be reasoned	Apply §3 Step 1: single verifiable answer?	Route to code or prose	PoT premise: decouple compute from reasoning [arXiv 2211.12588]
2	Decompose mixed steps	Step has both a number and a narrative	Split into computable sub-step (code) + judgment sub-step (prose)	Two routed sub-steps	Mirrors mixed-content split in [[agentsop-output-format-by-model]] §5 Case B
3	Sandbox choice	Decided to emit code	Pick interpreter by trust + capability: DSPy PoT (Python sandbox), OpenAI Code Interpreter (managed container), Anthropic code-exec (VM), LangChain `PythonREPLTool` (unsandboxed, in-process)	Chosen runtime	DSPy modules [dspy.ai/learn/programming/modules/]; LangChain PythonREPLTool docs
4	Result-back-into-LM	Code produced a value	Inject stdout/return value into the next LM turn so the model narrates/uses it	LM-consumed result	PoT design: code computes, LM contextualizes [arXiv 2211.12588]
5	Error retry (bounded)	Code raised a traceback	Feed error to LM, regenerate, re-run; cap at `max_iters` (~3) then fall back	Fixed code or graceful fallback	DSPy `ProgramOfThought` `max_iters`; see [[agentsop-test-fix-loop]]
6	Precision escalation	Prose answer involves multi-step arithmetic	Re-route the arithmetic to code even if prose started it	Code-verified number	"LMs are unreliable calculators" — PAL [arXiv 2211.10435]
7	Over-coding veto	About to sandbox a judgment task	Stop: no deterministic core → reasoning, not code	Prose reasoning, no sandbox call	§5 Case B; avoids wasted round-trip

In DSPy terms, op #1 is exactly the choice between dspy.ChainOfThought (prose reasoning) and dspy.ProgramOfThought (emit+run) for a signature — the dspy skill lists the modules but this overlay supplies the when.

5. 困境决策案例 (Dilemma cases)

Case A — Arithmetic in prose hallucinates → PoT fixes it (under-coding)

Trigger: A finance-summary agent step: "Given these 14 line items, compute the total, the 8.25% tax, and the grand total." The agent is a dspy.ChainOfThought module emitting prose.

Constraints:

14 multi-digit addends + a percentage + a sum-of-sums. Well outside reliable mental-math range.
The number flows into an invoice — wrong is expensive and silent.
A Python sandbox is available.

Decision steps:

Gate (§3 Step 1): "Is there a single verifiable answer?" Yes — the total is a fact, not a judgment. → computable.
Recognize the under-coding failure mode. Prose chain-of-thought will emit a plausible running sum that is frequently off by some digits, and no exception will fire. This is the silent failure the gate exists to prevent.
Switch the module from ChainOfThought to ProgramOfThought. The model now emits subtotal = sum([...]); tax = round(subtotal * 0.0825, 2); total = subtotal + tax and the interpreter computes it exactly.
Feed the result back (op #4): the LM receives total = 4,217.93 and narrates the invoice line. Code computed; LM contextualized.
Serialize per [[agentsop-output-format-by-model]]: the generated code goes in a fenced block, not a JSON program field — code-in-JSON would degrade it.

Outcome: The arithmetic is now deterministic and auditable. PoT-style code execution is the documented fix for exactly this class of error [arXiv 2211.12588, arXiv 2211.10435].

Extractable operation: Multi-step arithmetic in prose is a smell. Re-route it to code (op #6).

Case B — Over-coding a judgment task wastes a sandbox round-trip (over-coding)

Trigger: A support-triage agent step: "Read this customer message and decide whether the tone is hostile, neutral, or warm." An over-eager engineer wires it through ProgramOfThought because "code is more reliable."

Constraints:

"Tone" has no single verifiable answer — it is a judgment.
The sandbox round-trip adds latency + cost.
Forcing it into code produces something like if "!!!" in msg: tone = "hostile" — a brittle rule that encodes a subjective rubric as if it were a formula, and is worse than the model's native judgment.

Decision steps:

Gate (§3 Step 1): "Single verifiable answer a program could check?" No — tone is judgment. → reason in prose. Stop.
Invoke the over-coding veto (op #7). There is no deterministic core to delegate to an oracle. Code adds no determinism — it just relocates the same judgment into worse, hard-coded heuristics.
Cost it out. The sandbox call adds a full round-trip (and a code-gen + parse step) for zero precision gain. This is pure waste — the loud, cheap failure mode.
Keep it as ChainOfThought (or Predict). The LM is the right engine for judgment.
Counter-check: if a later spec adds "and count how many exclamation marks" — that sub-step is computable; decompose (op #2) and route the count to code, the tone to prose.

Outcome: No sandbox call. The judgment stays where judgment belongs. Over-coding is a real and common anti-pattern: not everything benefits from an interpreter, only things with a verifiable deterministic core.

Extractable operation: No verifiable answer → no code. Veto the sandbox round-trip (op #7).

Case C — Mixed step: summary + total (decompose)

Trigger: "Summarize this quarter's sales narrative and give me the exact total revenue."

Constraints: One sentence contains both a judgment (summary) and a computation (total).

Decision steps:

Gate → mixed. Decompose (op #2).
Route the total to ProgramOfThought: code sums the figures, interpreter verifies.
Route the summary to ChainOfThought: prose synthesis, no oracle exists.
Feed the code result back into the prose step so the narrative cites the exact, verified total.

Outcome: Each sub-step uses its correct engine. This is the execution-decision analogue of the mixed-content two-pass pattern in [[agentsop-output-format-by-model]] §5 Case B.

Extractable operation: One task ≠ one strategy. Gate per step, decompose mixed steps.

6. 反模式与边界 (Anti-patterns & boundaries)

Anti-patterns

Code for everything ("code is always more reliable"). False. Code is more reliable only where a deterministic, verifiable answer exists. For judgment tasks, code just hard-codes a subjective rubric and adds a wasted round-trip (Case B). The interpreter is an oracle only for computable questions.
Reasoning for arithmetic ("the model can just add it"). The headline failure. LMs are unreliable calculators; multi-step prose arithmetic hallucinates plausible wrong numbers, silently (Case A). Any non-trivial computation → code [arXiv 2211.12588].
Emitting code but never feeding the result back. PoT is "code computes, LM contextualizes." If the value never re-enters the LM turn, you have a dangling computation the agent can't use or narrate (op #4).
Unbounded error-retry loops. On a traceback, regenerate-and-rerun is correct — but cap it (max_iters ≈ 3) and fall back to prose or surface the failure. Infinite code-repair loops burn cost. See [[agentsop-test-fix-loop]].
Nesting generated code inside JSON sub-structure. The execution decision says emit code; [[agentsop-output-format-by-model]] says emit it as a fenced block / single string, never as nested JSON — code-in-JSON degrades the code (Aider 61%→20%). Pass both gates.
Treating LangChain PythonREPLTool as a safe sandbox. It runs in-process, unsandboxed. Fine for trusted self-authored code; hostile to untrusted input. Use a real sandbox (OpenAI Code Interpreter container, Anthropic code-exec VM, DSPy's interpreter) when inputs are untrusted.
Over-decomposing trivial computation. A 3-item count or single-digit add does not need a sandbox round-trip; the tax exceeds the benefit. The gate is for non-trivial computation.

Boundaries (when this gate doesn't fire)

No sandbox available and determinism is soft. Sometimes you must reason in prose because there is no interpreter; accept the precision risk and flag it.
The model is overwhelmingly capable for a tiny computation. Frontier models reliably handle small arithmetic; the round-trip isn't worth it below a complexity threshold.
The task is one-shot exploration with no downstream consumer of the precise value.
Latency-critical paths where a sandbox round-trip violates a hard SLA and an approximate answer is acceptable — a deliberate, documented tradeoff, not a default.

7. 跨框架对照 (Ecosystem cross-reference)

How major frameworks expose the emit-code-vs-reason mechanism. This skill operates at the decision layer; each framework supplies the mechanism.

DSPy `ProgramOfThought` [dspy.ai/learn/programming/modules/]

The canonical declarative version. A signature compiled with dspy.ProgramOfThought(Sig) makes the LM emit Python, runs it in an interpreter sandbox, and feeds the result back — with bounded retry on error (max_iters). The sibling dspy skill lists ChainOfThought vs ProgramOfThought as module choices but does not give the decision rubric; this overlay is that rubric. Use ProgramOfThought exactly when §3 Step 1 returns "computable."

OpenAI Code Interpreter / Assistants code tool

A managed container that the model can write Python into and execute, with files and state persisted across turns. Heavier and stateful — good for data-analysis sessions (load CSV, compute, plot). Same gate applies: route computable steps in, keep judgment in chat.

Anthropic code-execution tool

A sandboxed VM exposed as a tool; the model emits code, it runs, results return to the conversation. First-party, sandboxed — safe for untrusted inputs. The execution-decision gate maps directly: offer the tool, but the model/agent should only invoke it for computable steps.

LangChain `PythonREPLTool`

A tool wrapping a Python REPL. Runs in-process and is unsandboxed — powerful and dangerous. Use only for trusted, self-authored computation; never expose it to untrusted input without an external sandbox. The decision rubric is identical; the safety profile is worst-in-class.

Cross-framework summary

Framework            | Mechanism                | Sandbox     | Result-back
---------------------|--------------------------|-------------|------------
DSPy PoT             | ProgramOfThought module  | Python intp | automatic (max_iters)
OpenAI Code Interp.  | Assistants code tool     | container   | persisted state
Anthropic code-exec  | code-execution tool      | VM (safe)   | into conversation
LangChain            | PythonREPLTool           | NONE (proc) | manual wiring

Every framework can be mis-invoked — pointed at a judgment task (over-code) or skipped for a computation (under-code). This overlay is the gate that decides invocation, regardless of which mechanism is underneath.

Quick reference card

┌──────────────────────────────────────────────────────────────────────┐
│                  EMIT-CODE-VS-REASON DECISION CARD                   │
├──────────────────────────────────────────────────────────────────────┤
│ Single verifiable answer a program could check?                      │
│   YES → EMIT CODE → sandbox → run → feed result back into LM         │
│   NO  → REASON IN PROSE (no oracle exists for judgment)              │
│   MIXED → decompose; route each sub-step independently               │
├──────────────────────────────────────────────────────────────────────┤
│ Arithmetic / parse / sort / count / regex / symbolic → CODE          │
│ Tone / quality / summary / design / synthesis        → PROSE         │
│ "summary AND total"                                  → SPLIT         │
├──────────────────────────────────────────────────────────────────────┤
│ LMs are unreliable calculators but reliable coders.                  │
│ Under-coding fails SILENTLY (wrong plausible number).                │
│ Over-coding fails LOUDLY+CHEAPLY (wasted sandbox round-trip).        │
│ When genuinely uncertain AND a verifiable core exists → lean CODE.   │
├──────────────────────────────────────────────────────────────────────┤
│ NEVER:                                                               │
│  • code a judgment task (over-coding veto)                          │
│  • do multi-step arithmetic in prose (under-coding)                 │
│  • forget to feed the code result back into the LM                  │
│  • nest generated code in JSON (see [[agentsop-output-format-by-model]])     │
│  • loop code-repair unbounded (see [[agentsop-test-fix-loop]])               │
│  • trust LangChain PythonREPLTool on untrusted input (unsandboxed)  │
└──────────────────────────────────────────────────────────────────────┘

引用源 (Citations)

Primary anchors:

Program of Thoughts (PoT) — Chen et al., arXiv 2211.12588 — decouple computation from reasoning; LM writes code, interpreter computes.
PAL: Program-aided Language Models — Gao et al., arXiv 2211.10435 — LM as coder, runtime as deterministic solver.
DSPy ProgramOfThought module — [dspy.ai/learn/programming/modules/] — emit+run+retry mechanism.

Framework / API docs:

OpenAI Code Interpreter / Assistants code tool — [platform.openai.com/docs/assistants/tools/code-interpreter]
Anthropic code-execution tool — [docs.anthropic.com/en/docs/agents-and-tools/tool-use/code-execution-tool]
LangChain PythonREPLTool — [python.langchain.com/docs/integrations/tools/python] (unsandboxed; in-process).

Companion / overlaid skills:

dspy-sop-skill/SKILL.md — ships ProgramOfThought but not this decision rubric (the gap this overlay fills).
d-output-format-by-model-skill/SKILL.md — sibling: once you emit code, how to serialize it (PoT for math/parse; code never nested in JSON). Cross-linked as [[agentsop-output-format-by-model]].
test-fix-loop — the execute → error → retry loop this skill defers to for bounded code repair. Cross-linked as [[agentsop-test-fix-loop]].

When it activates

Automatically, when a task matches. Use when designing or debugging an agent step that does arithmetic/parsing/data transforms, when prose reasoning hallucinates a computation (under-coding), or when a sandbox round- trip is wasted on a judgment task (over-coding). Search keyw

About this skill

Automatic

AI Engineer pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

agentsope / skillalchemy

skills/agentsop-code-execution-decision/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1Bundled 1 file

Illustrative Want to see what agentsope/agentsop-code-execution-decision would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Related skills

Skills that do similar work, with their headline benchmark lift.

From the same source and nearby in the catalogue.

Ready to supercharge your AI agents?

SkillScore

Safety

What it changes

Track record

Use this skill

Code-Execution Decision — emit-code-vs-reason

1. 何时激活 (When to activate)

2. 核心心智模型 (Core mental model)

3. SOP 工作流 (Decision workflow)

Step 1 — Classify the step: deterministic-computable vs judgment

Step 2 — If computable, emit + run code

Step 3 — Sandbox → run → feed result back into the LM

4. 操作模型 (Operations: Trigger → Action → Output → Evidence)

5. 困境决策案例 (Dilemma cases)

Case A — Arithmetic in prose hallucinates → PoT fixes it (under-coding)

Case B — Over-coding a judgment task wastes a sandbox round-trip (over-coding)

Case C — Mixed step: summary + total (decompose)

6. 反模式与边界 (Anti-patterns & boundaries)

Anti-patterns

Boundaries (when this gate doesn't fire)

7. 跨框架对照 (Ecosystem cross-reference)

DSPy ProgramOfThought [dspy.ai/learn/programming/modules/]

OpenAI Code Interpreter / Assistants code tool

Anthropic code-execution tool

LangChain PythonREPLTool

Cross-framework summary

Quick reference card

引用源 (Citations)

When it activates

SkillSafety

Source

Related skills

More in the registry

DSPy `ProgramOfThought` [dspy.ai/learn/programming/modules/]

LangChain `PythonREPLTool`