Ready to supercharge your AI agents?

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

Organic Chemistry Reasoning Guide · mims-harvard | DecimalAI

Scienceby mims-harvard

Organic Chemistry Reasoning Guide

Organic chemistry reasoning guide for reaction product prediction, mechanism analysis (electrophilic/nucleophilic substitution, addition, elimination, pericyclic, radical), and spectroscopy interpretation (1H/13C NMR, IR, MS). Reasons from first principles (electron flow, kinetic vs thermodynamic) rather than pattern-matching named reactions.

Never made a case worseMaintained

SkillScore

SkillScore83/100

Test results 83

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

79%83%of test cases passed +4 pts

Without skill19 of 24

With skill20 of 24

1 fixed19 passed either way4 still failing0 broke

Model 8/7/2026v1 · latest is v223 comparable

This benchmark measured an earlier version of the skill.

A newer attempt on gemini-3.8-flash on 11/1/2026 did not produce a publishable comparison (graded differently). The snapshot above is the previously measured run.

Track record

Agent runs: 15; All time

See one task, with and without the skill

One test taskCalculate the zero-point energy and the first excited state energy in Joules for a 1D harmonic oscillator with a fundamental vibrational frequency of 3000 cm^-1. Show the formulas used.

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 mims-harvard-tooluniverse-organic-chemistry
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("mims-harvard-tooluniverse-organic-chemistry")
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 mims-harvard-tooluniverse-organic-chemistry

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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md https://app.decimal.ai/s/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

This gets SKILL.md only. The skill's 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/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/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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("mims-harvard-tooluniverse-organic-chemistry")
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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/mims-harvard-tooluniverse-organic-chemistry/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 7 other files are 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 mims-harvard-tooluniverse-organic-chemistry

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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md https://app.decimal.ai/s/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

This gets SKILL.md only. The skill's 7 other files are 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/mims-harvard-tooluniverse-organic-chemistry/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/mims-harvard-tooluniverse-organic-chemistry.json

This gets SKILL.md only. The skill's 7 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v2 (v1 tested 8/7/2026)
Source: mims-harvard/ToolUniverse ↗
License: Apache-2.0

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

Organic Chemistry Reasoning Guide

This skill teaches how to think through organic chemistry problems, not what to memorize.

1. Predicting Reaction Products — A Reasoning Process

Do NOT pattern-match to a named reaction. Instead, reason from first principles:

Step 1 — Find the most reactive sites

Look at each molecule and ask: where is the electron density?

Electron-rich sites (nucleophiles): lone pairs on N, O, S; pi bonds (C=C, aromatics); carbanions; enolates
Electron-poor sites (electrophiles): carbons bonded to electronegative atoms (C=O, C-X); carbocations; atoms with empty orbitals

The reaction will happen where the strongest nucleophile meets the strongest electrophile.

Step 2 — What kind of bond-making/breaking is this?

Three fundamental categories:

Ionic (polar): electrons move in pairs. One species donates electrons, the other accepts. Most common in solution chemistry.
Radical: electrons move one at a time. Look for: heat + peroxides, UV light (hv), NBS, or radical initiators.
Pericyclic: electrons move in a concerted cyclic transition state. Look for: heat or light with no other reagents, dienes + dienophiles, sigmatropic rearrangements.

Step 3 — Where do the electrons flow?

Draw the arrow from nucleophile to electrophile. Ask:

Which atom has the best orbital overlap?
Is backside attack required (SN2) or can the nucleophile approach from either face (SN1)?
For additions to C=C: which carbon gets the electrophile? (Markovnikov = more substituted carbocation intermediate; anti-Markovnikov = radical or hydroboration)
For additions to C=O: the nucleophile attacks carbon (Burgi-Dunitz angle ~107 degrees).
For E2 eliminations: base size determines regiochemistry. Bulky bases (KOtBu, LDA, DBU) favor the Hofmann product (less substituted alkene) because they abstract the less sterically hindered proton. Small bases (NaOEt, KOH, NaOMe) favor the Zaitsev product (more substituted, more stable alkene). Always check base size before predicting regiochemistry.
For ion tracking in multi-step syntheses: write out the complete ionic equation at each step. Track every cation and anion through each transformation — identify what dissolves, what precipitates, what complexes form, and what remains in solution. Verify at each step: are all ions accounted for? Does the mass balance? Use solubility rules (Ksp) and complex stability constants (Kf) to predict which species persist.

Step 4 — What is the driving force?

Every reaction needs a thermodynamic reason to proceed:

Leaving group departure: good leaving groups (I > Br > Cl > F; tosylate, triflate) stabilize themselves after leaving
Ring strain relief: epoxide opening, cyclopropane opening
Aromaticity gain: tautomerization to aromatic form, elimination to form conjugation
Strong bond formation: C-O, C-F, H-O bonds are strong; reactions that form them are favored
Gas evolution: loss of CO2, N2, or H2O drives equilibrium forward

Step 5 — Sanity-check your product

Before reporting an answer:

Are all valences correct? (C=4, N=3, O=2, H=1, halogens=1)
Does the molecular formula balance? Count every atom on both sides.
Is the product thermodynamically reasonable? (Don't propose anti-aromatic products or strained small rings without justification)
Does degrees of unsaturation make sense? (DoU = (2C + 2 + N - H - X) / 2; a benzene ring = 4 DoU)

Named Reaction Decision Tree

Identify the reaction type from reagents/conditions, then apply its product topology:

Reagent/Condition Pattern	Reaction Type	Product Logic
RMgBr (or RLi) + ArX + Pd or Ni catalyst	Kumada coupling	R replaces X on Ar; excess RMgBr replaces ALL X
Ph3P=CHR + aldehyde/ketone	Wittig	C=C replaces C=O; unstabilized ylide → Z-alkene; stabilized (ester/CN) → E-alkene
Sulfoxide + strong electrophile (Tf2O, Ac2O)	Pummerer rearrangement	S-oxidation state drops; alpha-carbon gets new bond to nucleophile/leaving group
Allyl vinyl ether heated (or 1,5-dien-3-ol + base)	[3,3]-sigmatropic (oxy-Cope/Claisen)	Redraw 6-membered chair TS; new sigma bond at 1,6 positions; old 3,4 bond breaks
Diene + dienophile, heat	Diels-Alder	[4+2] cycloaddition; endo rule for stereochemistry; cis dienophile substituents stay cis
ArX + excess organometallic (no catalyst)	Nucleophilic aromatic substitution or benzyne	Each X replaced by nucleophile; count equivalents to determine degree of substitution

Retrosynthetic Analysis

Work backward from product to starting materials by identifying how key bonds were formed:

Robinson Annulation (product is a 2-cyclohexenone fused or substituted system):
- Disconnect the C-C bond alpha to the enone carbonyl → two fragments: a 1,3-dicarbonyl compound + methyl vinyl ketone (or equivalent Michael acceptor)
- The 1,3-dicarbonyl (e.g., ethyl 2-methyl-6-oxocyclohexanecarboxylate) is the starting material
- Mechanism: Michael addition then intramolecular aldol condensation
Intramolecular Aldol (fused ring from KOH/base on an open-chain substrate):
- Product is a fused bicyclic enone → starting material is a 1,5-diketone (or 1,n-diketone)
- Carbon counting rule: product ring size = carbons between the two carbonyls + 1
- Example: hexahydronaphthalen-2-one from KOH → start with 2-(3-oxopentyl)cyclohexanone, not a butyl chain
General disconnection heuristic — identify the new bond, then choose the right reaction:
- New C-C bonds: aldol, Claisen, Michael, Wittig, Grignard, Diels-Alder
- New C-O bonds: epoxidation, hydration, oxidation
- New C-N bonds: reductive amination, Gabriel synthesis, amide coupling

ML-Based Reaction/Retrosynthesis Prediction (Cross-Check, Not a Replacement)

RXNChemistry_predict_reaction (forward prediction: reactants → most likely product, IBM RXN for Chemistry ML model) and RXNChemistry_predict_retrosynthesis (target → candidate synthetic routes) can cross-check a from-first-principles answer, especially for reactions outside the named-reaction decision tree above. Reason through the mechanism first per the steps in this skill — these are a sanity check on a proposed product/route, not a substitute for understanding why it forms.

Requires RXN4CHEMISTRY_API_KEY — verified live: with no key set, neither tool appears in tu list/loads at all (the same intentional key-gating pattern as Addgene/USPTO elsewhere in ToolUniverse); set the key to enable, don't treat their absence as a bug. Both operations are asynchronous (poll via poll_interval/max_wait_time params) since the underlying model runs as a queued job, not a synchronous API call.

Product Prediction Strategy (Stepwise)

Reactive sites: Mark every nucleophilic and electrophilic center in all reactants
Reaction type: Match reagent/condition pattern to the decision tree above
Bond changes: Explicitly list which bonds break and which form (write them out)
Regiochemistry: Markovnikov (cation stability) vs anti-Markovnikov (radical/BH3); for ArX, each leaving group is an independent reactive site
Stereochemistry: Wittig E/Z from ylide type; SN2 inversion; [3,3] suprafacial-suprafacial (chair TS); Pummerer retains sulfur connectivity
Name the product: Convert the final structure to IUPAC; for biaryl systems count the phenyl groups systematically (e.g., three linked phenyls = terphenyl, not triphenylbenzene)

Structure Determination from Reactions

When a hydrocarbon reacts with Br2 under "normal conditions" (no UV, no catalyst): this is ionic ADDITION (to a strained ring or C=C), not radical substitution
If the product has exactly 1 Br per carbon that reacted: suspect ring-opening of a strained cyclopropane or addition across C=C
Strained 3-membered rings with Br2 undergo ionic ring-opening, not radical substitution
When a metal dissolves in a solution of its own higher-valent salt, the product is the intermediate oxidation state (comproportionation), not a metal displacement

Combustion Analysis Strategy

Preferred: use MolecularFormula_analyze tool (via MCP/SDK) with sample_g, CO2_g, H2O_g, and molar_mass parameters. Fallback: run molecular_formula.py directly.
Preferred: use DegreesOfUnsaturation_calculate tool with formula parameter. Fallback: run degrees_of_unsaturation.py --formula <result>.
Match DoU + any functional group test results (Br2 decolorization, KMnO4, etc.) to narrow the structure

2. Spectroscopy Interpretation — A Problem-Solving Strategy

Do NOT look up tables of chemical shifts. Instead, follow this systematic approach:

Start with molecular formula

If given, calculate degrees of unsaturation first. This immediately tells you:

DoU = 0: no rings or double bonds (saturated, open chain)
DoU = 1: one ring OR one double bond
DoU = 4: almost certainly a benzene ring (3 double bonds + 1 ring)
DoU = 2: could be a triple bond, or two double bonds, or a ring + double bond

This single number eliminates huge categories of structures before you look at any spectrum.

IR: Look for the obvious peaks first

Do not try to assign every peak. Instead, ask three questions:

Is there a broad absorption around 2500-3500? If yes: O-H (very broad) or N-H (sharper, sometimes two peaks for primary amine).
Is there a strong sharp peak around 1650-1750? If yes: C=O. Its exact position tells you the type (ester ~1735, ketone ~1715, acid ~1710, amide ~1680).
Is there a peak around 2100-2300? If yes: triple bond (C-triple-C or C-triple-N) or cumulated double bond.

If you can answer these three questions, you have identified the major functional groups. Everything else is confirmatory.

1H NMR: Three pieces of information per signal

Every signal tells you three things — use all of them:

Chemical shift (where): How deshielded is this proton? Near electronegative atoms = downfield (higher ppm). Aromatic protons are 6.5-8.5 ppm; alkyl are 0.8-1.7 ppm; adjacent to C=O are 2.0-2.7 ppm.
Multiplicity (splitting pattern): How many neighboring protons does this one have? n neighbors = n+1 peaks. A triplet means 2 adjacent H; a quartet means 3 adjacent H. This tells you about connectivity.
Integration (area): How many protons does this signal represent? Ratios matter more than absolute values. A 3:2:1 integration ratio on a C3 compound strongly suggests CH3, CH2, CH.

13C NMR: Count the symmetry

Count the number of distinct carbon signals. Compare to the molecular formula.
Fewer signals than carbons = the molecule has symmetry (equivalent carbons).
Carbonyl carbons appear above 160 ppm. Aromatic/alkene carbons: 100-160 ppm. sp3 carbons: 0-90 ppm.

The combination strategy

No single spectrum gives the answer. Combine them:

Molecular formula gives DoU (structural constraints)
IR identifies functional group classes (C=O? O-H? N-H?)
13C tells you how many unique carbon environments exist
1H tells you connectivity (through splitting) and hydrogen count (through integration)
Propose a structure that is consistent with ALL data
Verify: does your proposed structure predict every observed signal?

3. Stereochemistry — How to Think in 3D

Finding stereocenters

A stereocenter is a carbon with four DIFFERENT substituents. To find them:

Look at each sp3 carbon in the molecule
Check: are all four groups attached to it different? (Trace each path outward — if two paths become identical at any point, those groups are the same)
Ring carbons count too — the two "ring paths" going in opposite directions around the ring are different substituents if they differ at any point

Assigning R/S

Rank the four substituents by CIP priority (higher atomic number = higher priority)
For ties: move to the next atom along the chain and compare again
Orient the molecule so the lowest-priority group (#4) points AWAY from you
Read priorities 1-2-3: clockwise = R, counterclockwise = S
Shortcut: if #4 is already pointing away in a drawing, just read the arrangement directly. If #4 points toward you, read the arrangement and FLIP the answer.

Predicting stereochemical outcomes of reactions

Ask: what is the mechanism?

SN2: inversion at the stereocenter (backside attack). Always.
SN1: racemization (planar carbocation intermediate attacked from both faces). Expect ~50:50 or slight preference.
E2: requires anti-periplanar geometry (H and leaving group 180 degrees apart). This constrains which alkene geometry forms.
Addition to alkenes: syn addition (hydroboration, catalytic hydrogenation) vs anti addition (bromine, epoxidation then opening)
Addition to carbonyls with alpha-stereocenter: Felkin-Anh model — nucleophile attacks anti to the largest substituent on the alpha carbon

Relationship between stereoisomers

Enantiomers: non-superimposable mirror images. Same physical properties except optical rotation.
Diastereomers: stereoisomers that are NOT mirror images. Different physical properties — can be separated.
Meso compounds: have stereocenters but an internal mirror plane makes them achiral overall.

4. Bundled Scripts

These ready-to-run scripts live in skills/tooluniverse-organic-chemistry/scripts/. Use them via the Bash tool instead of computing by hand — they validate inputs and print verification lines.

`chemistry_facts.py` — Chemical and physical facts reference

A lookup tool for facts that are easy to mis-remember. Query this script instead of guessing. Three categories: allotropes, point_group, common_reagents. Run with no extra args to list all entries in a category.

python chemistry_facts.py --type allotropes --element P
python chemistry_facts.py --type point_group --molecule "benzene"
python chemistry_facts.py --type common_reagents --reagent "LiAlH4"
python chemistry_facts.py --type allotropes          # list all elements

Coverage: Allotropes for P, C, S, O. Point groups for 15 molecules (water C2v through H2 D∞h). 24 reagents (LiAlH4 through DMDO) with full name, function, selectivity, stereochemistry, and KEY DISTINCTION notes.

When to use: Any question about allotropes, molecular symmetry/point groups, or reagent selectivity. Always query before reporting reagent facts.

`crystal_validator.py` — Crystal structure density validator

Preferred: use CrystalStructure_validate tool (via MCP/SDK) with unit cell parameters (a, b, c, alpha, beta, gamma, Z, MW, density). Fallback: run crystal_validator.py directly.

Validates crystal structure data by computing theoretical density from unit cell parameters and comparing against a reported value. Flags inconsistencies that indicate errors in published datasets.

Supports all seven crystal systems (cubic, tetragonal, orthorhombic, hexagonal, trigonal, monoclinic, triclinic).

python crystal_validator.py --a 5.43 --b 5.43 --c 5.43 --Z 8 --MW 28.085 --density 2.329
python crystal_validator.py --datasets datasets.json   # batch mode: JSON array of datasets

Angles default to 90deg. Output: crystal system, volume, calculated density, %error, verdict (OK/WARNING/MISMATCH). Batch mode compares multiple datasets to find the outlier. Density formula: d = (Z * MW) / (V * Na).

When to use: Crystal structure verification, density consistency checks, finding erroneous datasets.

`stereochem_tracker.py` — Stereochemistry through reaction sequences

Tracks R/S configuration at a stereocenter through a series of reactions, predicting the stereochemical outcome at each step.

python stereochem_tracker.py --start R --reactions "SN2, oxidation, SN1"

Supported reactions: SN2 (inversion), SN1 (racemization), E2/E1 (elimination), reduction/oxidation/hydrogenation/hydroboration/epoxidation (retention), mitsunobu (inversion), double_inversion (retention), racemization/epimerization/enolization (racemization).

When to use: Multi-step synthesis chirality tracking, verifying net retention/inversion.

`smiles_verifier.py` — SMILES molecular property verifier (no RDKit)

Preferred: use SMILES_verify tool (via MCP/SDK) with smiles and optional constraint parameters (mw, heavy_atoms, valence_electrons, total_atoms, formal_charge). Fallback: run smiles_verifier.py directly.

Parses a SMILES string without external dependencies and computes molecular weight, heavy atom count, valence electron count, total atom count, and formal charge. Then optionally checks these against user-supplied constraints. Use this every time you design or propose a SMILES answer.

python smiles_verifier.py --smiles "CC(C)(C(=N)N)N=NC(C)(C)C(=N)N" --mw 198.15 --heavy_atoms 14 --valence_electrons 80

Constraint flags: --mw (tolerance 0.5), --heavy_atoms, --valence_electrons, --total_atoms, --formal_charge

When to use: Always verify SMILES answers before reporting. Catches MW, atom count, and electron count errors.

`degrees_of_unsaturation.py` — Degrees of unsaturation (DoU) calculator

Preferred: use DegreesOfUnsaturation_calculate tool (via MCP/SDK) with formula or individual atom count parameters. Fallback: run degrees_of_unsaturation.py directly.

Computes DoU = (2C + 2 + N − H − X) / 2 from a formula string or individual atom counts. Handles all halogens (F, Cl, Br, I) and notes that O and S do not affect DoU. Prints the full arithmetic, a structural interpretation, and a round-trip verification.

python degrees_of_unsaturation.py --formula C6H6
python degrees_of_unsaturation.py --C 10 --H 14 --O 2 --N 1

Output: atom counts, step-by-step formula, result, structural interpretation, non-integer DoU warning.

`molecular_formula.py` — Combustion analysis and formula analysis

Preferred: use MolecularFormula_analyze tool (via MCP/SDK) with combustion or formula parameters. Fallback: run molecular_formula.py directly.

Two modes in one script.

Mode 1: combustion analysis — --sample_g, --CO2_g, --H2O_g, optional --molar_mass Mode 2: formula analysis — --formula C6H6 (MW, DoU, elemental composition)

python molecular_formula.py --sample_g 0.2 --CO2_g 0.4874 --H2O_g 0.1998 --molar_mass 78
python molecular_formula.py --formula C6H6

5. Computational Procedures

ALWAYS EXECUTE as Python scripts using the Bash tool — do not compute mentally. Prefer the bundled scripts (molecular_formula.py, degrees_of_unsaturation.py) over writing code from scratch.

Key formulas (write and run Python when bundled scripts don't cover the case):

Combustion analysis: moles_C = CO2_g / 44.01; moles_H = H2O_g / 18.015 * 2; moles_O by mass difference. Ratio → GCD → empirical formula. Factor = MM / empirical_MM.
DoU: (2C + 2 + N - H - X) / 2. O and S do not affect the count.
Limiting reagent: equiv = mass / (MW * stoich_coeff) for each reagent; minimum equiv is limiting; yield = min_equiv * product_stoich * product_MW.

6. When to Reason vs When to Look Things Up

REASON through these (do not look up):

Reaction products and mechanisms
Spectroscopy interpretation (combine the data you have)
Stereochemical outcomes
Degrees of unsaturation
Whether a reaction is thermodynamically favorable

COMPUTE with bundled scripts for these:

Crystal structure density verification: Preferred: CrystalStructure_validate tool. Fallback: crystal_validator.py (never compute unit cell volumes by hand)
Stereochemistry tracking through multi-step synthesis: stereochem_tracker.py (tracks R/S through SN2, SN1, etc.)
Degrees of unsaturation: Preferred: DegreesOfUnsaturation_calculate tool. Fallback: degrees_of_unsaturation.py
Combustion analysis and formula analysis: Preferred: MolecularFormula_analyze tool. Fallback: molecular_formula.py
SMILES verification (MW, heavy atoms, valence electrons): Preferred: SMILES_verify tool. Fallback: smiles_verifier.py (always verify SMILES answers)

LOOK UP with tools for these:

Allotropes of an element (count, names, properties): chemistry_facts.py --type allotropes --element X
Molecular point group and symmetry elements: chemistry_facts.py --type point_group --molecule "name"
Reagent function, selectivity, or key distinction: chemistry_facts.py --type common_reagents --reagent "name"
Specific physical constants (boiling points, pKa values, solubility): PubChem_get_CID_by_compound_name then PubChem_get_compound_properties_by_CID
Systematic (IUPAC) name → structure: OPSIN_name_to_structure (param name) — deterministic parser returning SMILES/InChI/InChIKey, the go-to for turning a systematic name (e.g. "2-acetoxybenzoic acid") into a structure you can then SMILES_verify. Trade/trivial names return parsed=false → use PubChem_get_CID_by_compound_name instead.
Whether a compound exists and its canonical SMILES/InChI: PubChem_get_CID_by_compound_name or PubChem_get_compound_properties_by_CID
Drug-likeness, bioactivity data, assay results: ChEMBL_get_molecule or ChEMBL_search_molecules
Metabolite pathways and biological context: HMDB_get_metabolite

Crystal structure verification strategy:

Extract unit cell params, Z, MW, density for each dataset. Run crystal_validator.py --datasets datasets.json in batch. Largest %error = the error. Common errors: wrong Z, swapped params, angle transcription errors.

Molecular Design from Constraints (SMILES design):

Start from constraints (MW, heavy atoms, VE) to estimate formula
Build SMILES, then always verify: python smiles_verifier.py --smiles "..." --mw X --heavy_atoms Y --valence_electrons Z
If any constraint FAILs, revise and re-run until all pass

The verify-after-reasoning pattern:

Reason through the problem, arrive at a proposed answer
Verify known compounds with PubChem; verify numerical answers with Python computation
For numerical answers: ALWAYS compute with scripts before answering

When it activates

Automatically, when a task matches. Use for organic synthesis problems and mechanism explanations.

About this skill

Automatic

Tutoring & Education pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v8 — 10/29/2026

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

mims-harvard / ToolUniverse

plugins/tooluniverse/skills/tooluniverse-organic-chemistry/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 10/29/2026Version v2Bundled 7 files · scripts

Illustrative Want to see what mims-harvard/tooluniverse-organic-chemistry would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

tooluniverse-organic-chemistryOrganic chemistry reasoning guide for reaction product prediction, mechanism analysis (electrophilic/nucleophilic substitution, addition, elimination, pericyclic, radical), and spectroscopy interpretation (1H/13C NMR, IR, MS). Reasons from first principles (electron flow, kinetic vs thermodynamic) rather than pattern-matching named reactions. Use for organic synthesis problems and mechanism explanations.Near-identical description+12 pts pass-rate lift

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

Organic Chemistry Reasoning Guide

1. Predicting Reaction Products — A Reasoning Process

Step 1 — Find the most reactive sites

Step 2 — What kind of bond-making/breaking is this?

Step 3 — Where do the electrons flow?

Step 4 — What is the driving force?

Step 5 — Sanity-check your product

Named Reaction Decision Tree

Retrosynthetic Analysis

ML-Based Reaction/Retrosynthesis Prediction (Cross-Check, Not a Replacement)

Product Prediction Strategy (Stepwise)

Structure Determination from Reactions

Combustion Analysis Strategy

2. Spectroscopy Interpretation — A Problem-Solving Strategy

Start with molecular formula

IR: Look for the obvious peaks first

1H NMR: Three pieces of information per signal

13C NMR: Count the symmetry

The combination strategy

3. Stereochemistry — How to Think in 3D

Finding stereocenters

Assigning R/S

Predicting stereochemical outcomes of reactions

Relationship between stereoisomers

4. Bundled Scripts

chemistry_facts.py — Chemical and physical facts reference

crystal_validator.py — Crystal structure density validator

stereochem_tracker.py — Stereochemistry through reaction sequences

smiles_verifier.py — SMILES molecular property verifier (no RDKit)

degrees_of_unsaturation.py — Degrees of unsaturation (DoU) calculator

molecular_formula.py — Combustion analysis and formula analysis

5. Computational Procedures

6. When to Reason vs When to Look Things Up

REASON through these (do not look up):

COMPUTE with bundled scripts for these:

LOOK UP with tools for these:

Crystal structure verification strategy:

Molecular Design from Constraints (SMILES design):

The verify-after-reasoning pattern:

When it activates

SkillSafety

Source

Other skills for the same job

Related skills

More in the registry

`chemistry_facts.py` — Chemical and physical facts reference

`crystal_validator.py` — Crystal structure density validator

`stereochem_tracker.py` — Stereochemistry through reaction sequences

`smiles_verifier.py` — SMILES molecular property verifier (no RDKit)

`degrees_of_unsaturation.py` — Degrees of unsaturation (DoU) calculator

`molecular_formula.py` — Combustion analysis and formula analysis