Ready to supercharge your AI agents?

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

CSV Documentation Generator · leoyeai | DecimalAI

Documentationby leoyeai

CSV Documentation Generator

Use when generating computer system validation (CSV) documentation for pharmaceutical and medical device industries, including validation plans, URS, FS, IQ/OQ/PQ documents, and traceability matrices.

Never made a case worse

SkillScore

SkillScore84/100

Test results 84

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

16%84%of test cases passed +68 pts

Without skill4 of 25

With skill21 of 25

17 fixed4 passed either way4 still failing0 broke

Measured on gemini-3.6-flash8/8/2026v123 comparable

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskWe want to perform a sync between our URS template and `requirements.json`, but strictly enforce that updates only flow from the template file into the JSON file (one-way read from template). Which flag and value configure this specific sync direction?

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 leoyeai-csv-documentation-generator
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("leoyeai-csv-documentation-generator")
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 leoyeai-csv-documentation-generator

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/leoyeai-csv-documentation-generator/SKILL.md https://app.decimal.ai/s/leoyeai-csv-documentation-generator/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/leoyeai-csv-documentation-generator/SKILL.md

This gets SKILL.md only. The skill's 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/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/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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("leoyeai-csv-documentation-generator")
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/leoyeai-csv-documentation-generator/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/leoyeai-csv-documentation-generator/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 23 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 leoyeai-csv-documentation-generator

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/leoyeai-csv-documentation-generator/SKILL.md https://app.decimal.ai/s/leoyeai-csv-documentation-generator/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/leoyeai-csv-documentation-generator/SKILL.md

This gets SKILL.md only. The skill's 23 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/leoyeai-csv-documentation-generator/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/leoyeai-csv-documentation-generator.json

This gets SKILL.md only. The skill's 23 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 8/8/2026
Source: LeoYeAI/openclaw-master-skills ↗
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

CSV Documentation Generator

Generate Computerized System Validation (CSV) documentation for pharmaceutical and medical device industries. Supports bilingual (Chinese/English) templates and multiple output formats (Word/Excel).

When to Use

Use this skill when:

Creating validation documentation for GMP-regulated systems
Generating User Requirements Specification (URS) with regulatory compliance checks
Preparing IQ/OQ/PQ protocols for system qualification
Creating traceability matrices linking requirements to test cases
Conducting risk assessments for computerised systems
Need documentation templates that comply with GAMP 5, 21 CFR Part 11, EU Annex 11

Supported Systems

System Type	GAMP Category	Documents
EDC (Electronic Data Capture)	Category 4/5	All 12 documents
CTMS (Clinical Trial Management)	Category 4	VP, URS, FS, RA, IQ, OQ, RTM
LIMS (Laboratory Information Management)	Category 4/5	All 12 documents
MES (Manufacturing Execution System)	Category 4	All 12 documents
ERP (Enterprise Resource Planning)	Category 4	VP, URS, RA, IQ, OQ
SCADA/DCS	Category 2/3	VP, URS, IQ, OQ, PQ
Custom Software	Category 3/4	VP, URS, RA, IQ, OQ

Quick Start

# Navigate to skill directory
cd <SKILL_DIR>/csv-documentation-generator

# Run script - will automatically create virtual environment on first run
python3 scripts/generate.py vp --project "XX系统" --system "EDC v1.0" --output ./output/

# Note: If --category is not specified, interactive GAMP category selection will be prompted
# The script provides bilingual (Chinese/English) guidance based on GAMP 5 Second Edition

# Generate Validation Plan (with category specified)
python3 scripts/generate.py vp --project "XX系统" --system "EDC v1.0" --category 4 --output ./output/

# Generate full validation package
python3 scripts/generate.py all --project "XX系统" --system "MES" --category 4 --output ./validation/

Parse Requirements from Code (pre-generation)

# Navigate to skill directory
cd <SKILL_DIR>/csv-documentation-generator

# Parse source code for requirements (extracts @URS, @TEST, @FS, @TS markers)
python3 scripts/cli.py parse ./src

# For AI agents (non-interactive mode), use --auto-add:
python3 scripts/cli.py parse ./src --auto-add

# Review extracted requirements
cat requirements.json

Auto Environment Setup

This skill automatically handles environment setup:

First run: Automatically creates .venv virtual environment in skill directory
Dependencies: Automatically installs required packages (python-docx, openpyxl, etc.)
Subsequent runs: Use the already-created virtual environment

This ensures consistent behavior and avoids dependency conflicts.

Interactive GAMP Category Selection

If --category is not specified, the script will prompt with bilingual (Chinese/English) guidance:

GAMP 5 (Second Edition) Category Selection:
  [1] Infrastructure Software - 基础设施软件
  [2] Firmware - 固件
  [3] Commercial Off-The-Shelf (COTS) - Non-configured - 商用现货软件 (不可配置)
  [4] Configured COTS - 配置型 COTS 软件 (e.g., EDC, CTMS, LIMS, MES)
  [5] Custom / Critical Application - 定制/关键应用

AI agents should guide users to select the appropriate category based on GAMP 5 Second Edition principles.

Available Commands

Command	Description	Output Format
`generate.py vp`	Validation Plan	.docx
`generate.py urs`	User Requirements Specification	.docx
`generate.py fs`	Functional Specification	.docx
`generate.py ts`	Technical Specification	.docx
`generate.py ra`	Risk Assessment (FMEA)	.docx
`generate.py iq`	Installation Qualification	.docx
`generate.py oq`	Operational Qualification	.docx
`generate.py pq`	Performance Qualification	.docx
`generate.py rtm`	Traceability Matrix	.xlsx
`generate.py vsr`	Validation Summary Report	.docx
`generate.py checklist`	Validation Checklist	.xlsx
`generate.py test-case`	Test Case Template	.xlsx
`generate.py all`	Full validation package	.docx / .xlsx

Command Options

Option	Description	Required
`--project`	Project name	Yes
`--system`	System name and version	Yes
`--category`	GAMP category (1-5). If not specified, interactive selection will be prompted with bilingual GAMP 5 guidance.	No (will prompt)
`--bilingual`	Enable bilingual mode: 'true' or 'false' (default: true). When true, headers remain bilingual and content follows --language.	No
`--language`	Primary language for content: 'zh' (Chinese) or 'en' (English) (default: zh). Used when bilingual=false or for content in bilingual mode.	No
`--verbose` / `-v`	Show detailed progress information (default: simplified output)	No
`--output`	Output directory	Yes
`--format`	Output format: docx, xlsx, or both (default: both)	No

Output Format Standards

All generated documents follow consistent naming and formatting rules:

Document Naming Convention

Document Type	Format	Example
Validation Plan	`{TYPE}_{PROJECT}.{ext}`	VP_XX系统.docx
User Requirements	`{TYPE}_{PROJECT}.{ext}`	URS_XX系统.docx
Functional Specification	`{TYPE}_{PROJECT}.{ext}`	FS_XX系统.docx
Traceability Matrix	`RTM_{PROJECT}.{ext}`	RTM_XX系统.xlsx

Output Directory Structure

{output_dir}/
├── VP_项目名.docx
├── URS_项目名.docx
├── FS_项目名.docx
├── TS_项目名.docx
├── RA_项目名.docx
├── IQ_项目名.docx
├── OQ_项目名.docx
├── PQ_项目名.docx
├── VSR_项目名.docx
├── RTM_项目名.xlsx
└── CHECKLIST_项目名.xlsx

Document Styling Standards

Fonts:

Chinese text: SimSun (宋体)
English text: Arial

Colors:

Table headers: #2563EB (Royal Blue) with white text
Priority markers: [必须] (red), [应该] (amber), [可以] (green)
Test results: [通过] (green), [失败] (red), [待测] (amber)

Template Variables

Common variables used in templates:

Variable	Description	Example
`{PROJECT_NAME}`	Project name	临床试验系统
`{SYSTEM_NAME}`	System name	EDC v1.0
`{SYSTEM_VERSION}`	System version	1.0
`{GAMP_CATEGORY}`	GAMP category	4
`{DOC_ID}`	Document ID	URS-001
`{DATE}`	Document date	2024-01-01
`{AUTHOR}`	Document author	张三
`{REVIEWER}`	Reviewer name	李四
`{APPROVER}`	Approver name	王五

Regulatory Compliance

GAMP 5 Categories

Category	Description	Validation Approach
1	Operating System	Legacy
2	Firmware	Simplified
3	Commercial-off-the-shelf (COTS)	Risk-based
4	Configured COTS	Risk-based
5	Custom/Critical	Full validation

Key Regulatory Requirements

This skill includes compliance checks for:

21 CFR Part 11: Electronic Records and Signatures
- Audit trail requirements
- Electronic signature requirements
- System access controls
EU Annex 11: Computerised Systems
- Validation requirements
- Data integrity
- Change control
ALCOA+: Data Integrity
- Attributable
- Legible
- Contemporaneous
- Original
- Accurate (+ Complete, Consistent, Enduring)

Semantic Action Graph

This skill encapsulates the following semantic actions for GxP document generation:

Action Graph

parse_requirements(source_path) → RequirementsDB
    │
    ▼
generate_vp(context) → Document
    │ ──────────────────────► generate_urs(context) → Document
    │                                           │
    │ ◄── (URS traceability) ──────────────────┤
    │                                           ▼
    ├────────────────────────────────────► generate_fs(context) → Document
    │                                           │
    │ ◄── (FS technical specs) ─────────────────┤
    │                                           ▼
    └────────────────────────────────────► generate_ra(context) → Document
                                                    │
                                        ┌───────────┴───────────┐
                                        ▼                       ▼
                              generate_iq_oq_pq(context)   generate_rtm(context)
                                        │                       │
                                        └───────────┬───────────┘
                                                    ▼
                                        generate_vsr(context) → Document

Semantic Actions

Action	Description	Pre-conditions
`parse_requirements()`	Parse @URS, @FS, @TEST markers from source code	Source code available
`generate_vp()`	Create Validation Plan	Project name, system name, GAMP category
`generate_urs()`	Create User Requirements Specification	System context, GxP scope
`generate_fs()`	Create Functional Specification	URS requirements linked
`generate_ra()`	Create Risk Assessment	URS priorities, GxP scope
`generate_iq_oq_pq()`	Create Qualification Protocols	RA risk levels, TS config
`generate_rtm()`	Create Traceability Matrix	URS, FS, test cases linked
`generate_vsr()`	Create Validation Summary Report	All test results, deviations
`sync_bidirectional()`	Sync requirements ↔ templates	Templates and requirements.json exist
`run_compliance_check()`	Verify GxP compliance	Requirements and test results

Data Flow Semantics

Flow	Semantic Meaning
VP → URS	Business process analysis defines GxP scope for requirements
URS → FS	Functional requirements trace to design specifications
URS → RA	Requirement priority determines risk severity
RA → Test	Risk level scales testing rigor (IQ/OQ/PQ scope)
All → RTM	Complete traceability matrix for regulatory audit

Prompt Library

Critical Thinking Constraints and Content Fill Prompts are defined in prompts.md.

For GAMP 5 M12 Critical Thinking constraints, document generation triggering conditions, and content fill prompt templates, see prompts.md.

Reference Documents

Reference materials included in references/ folder:

File	Description
`gamp-5.md`	GAMP 5 quick reference guide
`21cfr-part11.md`	21 CFR Part 11 key requirements
`annex-11.md`	EU Annex 11 requirements
`data-integrity.md`	ALCOA+ data integrity principles

Example Templates

Complete fill examples are provided in templates/examples/ directory:

Example File	Description
`urs-example.md`	URS with complete requirement examples
`fs-example.md`	FS with traceability examples
`ra-example.md`	RA with hierarchical risk examples
`iq-example.md`	IQ with qualification check examples

AI agents SHOULD reference these examples for content style guidance when filling templates.

Bilingual Format

All templates support Smart Bilingual output controlled by --bilingual and --language flags:

`--bilingual`	`--language`	Behavior
`true` (default)	`zh`	Headers bilingual, content in Chinese
`true` (default)	`en`	Headers bilingual, content in English
`false`	`zh`	Pure Chinese output
`false`	`en`	Pure English output

Bilingual Template Format

Templates use dual-language format for headers:

## 1. 目的 / Purpose

本文档定义了... / This document defines...

Tables use dual-language headers:

中文	English
验证计划	Validation Plan
用户需求	User Requirements

Content cells use format like 中文 / English:

模块 / Module	描述 / Description
用户管理 / User Management	系统应... / System should...

Requirements Traceability

This skill supports GAMP 5 compliant requirements traceability across the entire validation lifecycle.

Automated Traceability Chain

代码注释 (@URS[module]) 
    ↓
requirements.json (需求解析)
    ↓
URS 模板 (自动同步章节)
    ↓
FS/TS 文档 (引用需求ID)
    ↓
测试用例 (关联需求ID)
    ↓
测试结果 (解析需求ID)
    ↓
RTM (追溯矩阵自动生成)

Code Comment Standards (AI Agent Must Follow)

When generating code, AI agents MUST include requirement markers for automatic traceability. Use the standardized @REQ format:

# @REQ URS-001 - 系统应支持基于角色的访问控制 (RBAC)
# @REQ URS-002 - 用户密码必须满足复杂度要求（至少8位）
# @FS FS-001
# @TS TS-001
# @TEST[OQ-UM-001] - 验证用户角色分配功能
# @RISK H  # High risk module

def assign_role(user_id: str, role: str) -> bool:
    '''
    @REQ URS-001 - Role-based access control required
    @TEST[OQ-UM-001] - Test role assignment
    '''
    # Implementation...

Standard comment formats:

Pattern	Example	Description
`// @REQ URS-xxx`	`// @REQ URS-001 - 系统应支持RBAC`	Requirement with ID and description
`# @REQ`	`# @REQ URS-001 - 描述`	Python-style comment
`/* @REQ */`	`/* @REQ URS-001 - 描述 */`	Multi-line comment
`// @TEST[type-id]`	`// @TEST[OQ-UM-001] - 验证`	Test case link
`// @FS`	`// @FS FS-001`	FS reference
`// @TS`	`// @TS TS-001`	TS reference
`// @RISK [H/M/L]`	`// @RISK H`	Risk level (High/Medium/Low)

Risk levels:

H (High): Security, compliance, electronic signature, audit trail related
M (Medium): Default for most requirements
L (Low): Simple features, documentation, reports

Standard Modules

Module ID	中文名	English Name	Test Prefix
`user_mgmt`	用户管理	User Management	UM
`audit_trail`	审计追踪	Audit Trail	AT
`data_mgmt`	数据管理	Data Management	DM
`business_func`	业务功能	Business Functions	BF
`reporting`	报告功能	Reporting	RP
`integration`	接口集成	Integration	INT
`security`	安全	Security	SEC
`compliance`	合规	Compliance	CMP

注意: 章节号（4.X）是动态分配的，基于模板中已有的章节。同步时会自动为新模块分配下一个可用编号。

System Prompt Integration (Manual Configuration Required)

⛔ This step is NOT automatic. You must manually add these rules to your agent's system prompt (AGENTS.md) if you want @REQ/@TEST annotations to apply across ALL skills.

To enable cross-skill annotation enforcement, add the following to your AGENTS.md or agent system prompt:

Code Annotation Rules (for GxP traceability):
- When writing code that implements requirements, add @REQ markers: # @REQ URS-xxx - description
- When writing test cases, add @TEST markers: # @TEST[OQ-UM-xxx] - description
- Mark high-risk code sections with @RISK H (security, compliance, audit trail related)
- Mark medium-risk code with @RISK M
- Mark low-risk code with @RISK L

Why this is needed: OpenClaw skills only execute when active. There is no mechanism to make instructions persist across skill switches except via the system prompt.

See STANDARDS.md for full documentation and standards/code-annotations.json for the standards file.

Test Case ID Format

Test case IDs follow the format: {IQ|OQ|PQ}-{ModulePrefix}-{Number}

Type	Example	Description
IQ	`IQ-UM-001`	Installation Qualification test
OQ	`OQ-UM-001`	Operational Qualification test
PQ	`PQ-UM-001`	Performance Qualification test

Auto-Sync Feature

Before generating documents, use --sync to automatically update templates with new requirements:

# Generate URS and sync requirements to template
python3 scripts/generate.py urs --sync \
  --project "临床系统" \
  --system "EDC v1.0" \
  --category 4 \
  --output ./validation/

# Generate full validation package with sync
python3 scripts/generate.py all --sync \
  --project "XX系统" \
  --system "MES" \
  --category 4 \
  --output ./validation/

What --sync does:

Reads requirements.json for all requirements
Groups requirements by module
Checks if each module section exists in template
Auto-appends new module sections if missing
Backs up original template before modification

Sync Template to Database

Use --sync-template-to-db to extract requirements from templates and add them to requirements.json:

# Extract requirements from urs.md template to requirements.json
python3 scripts/generate.py urs --sync-template-to-db \
  --project "临床系统" \
  --system "EDC v1.0" \
  --category 4 \
  --output ./validation/

What --sync-template-to-db does:

Reads the template (e.g., urs.md)
Extracts all requirements in | URS-xxx | description | format
Loads existing requirements.json (if exists)
Adds new requirements that don't already exist
Preserves existing requirements and test results

Interactive Mode

For interactive step-by-step document generation with confirmation prompts:

python3 scripts/generate.py all --interactive \
  --project "临床系统" \
  --system "EDC v1.0" \
  --category 4 \
  --output ./validation/

Interactive workflow:

[1/11] 解析代码注释...
  → 按 Enter 继续，'s' 跳过，'q' 退出:

[2/11] 同步需求到数据库...
  → 按 Enter 继续，'s' 跳过，'q' 退出:

[3/11] 生成 URS...
  → 按 Enter 继续，'s' 跳过，'q' 退出:
  ✓ ./output/URS_临床系统.docx
...

Options:

Enter - Execute this step
s - Skip this step
q - Quit (already generated files are preserved)

Compliance Check

Run GAMP 5 compliance validation to check requirements coverage and test coverage:

# Run compliance check (text output)
python3 scripts/generate.py check \
  --requirements requirements.json \
  --test-results test_results.json

# Run compliance check (JSON output for CI/CD)
python3 scripts/generate.py check \
  --requirements requirements.json \
  --test-results test_results.json \
  --output-format json

Compliance checks performed:

Requirement coverage (ensures all code requirements are documented)
High-risk module verification (IQ/OQ/PQ tests for critical modules)
Test coverage threshold (default 80%)

Exit codes:

0 - All checks passed
1 - Warnings present
2 - Errors found

Incremental Update (Smart Rebuild)

Use --diff-only to skip regeneration when requirements haven't changed:

# Only regenerate RTM if requirements changed
python3 scripts/generate.py rtm --diff-only \
  --project "临床系统" \
  --system "EDC v1.0" \
  --category 4 \
  --output ./validation/

How it works:

Computes SHA256 hash of requirements after each generation
Stores hash in .requirements.hash in output directory
On subsequent runs, compares current hash against stored hash
Skips generation if hash matches (requirements unchanged)

Git Hooks (Automated Compliance)

Install post-commit hooks to automatically run compliance checks after code commits:

# Install hooks locally (per-project)
./scripts/git-hooks/install.sh --local

# Install hooks globally (all projects)
./scripts/git-hooks/install.sh --global

# Uninstall hooks
./scripts/git-hooks/install.sh --uninstall

What the hook does:

Detects commits with code or requirements changes
Runs compliance check (non-blocking)
Outputs warnings to stderr without blocking commit

Bidirectional Sync

Use --sync for bidirectional sync between template and requirements.json:

# Full bidirectional sync (both directions)
python3 scripts/generate.py urs --sync \
  --project "临床系统" \
  --system "EDC v1.0" \
  --category 4

# Only sync template → JSON
python3 scripts/generate.py urs --sync --sync-direction to-json

# Only sync JSON → template
python3 scripts/generate.py urs --sync --sync-direction to-template

Conflict resolution:

# Keep template descriptions on conflict
python3 scripts/generate.py urs --sync --conflict-resolution template

# Keep JSON descriptions on conflict
python3 scripts/generate.py urs --sync --conflict-resolution json

# Keep newer version based on updated_at
python3 scripts/generate.py urs --sync --conflict-resolution newer

Section numbering behavior:

Sync automatically creates module sections with ### 4.X headers
Section numbers are dynamically assigned based on existing sections in the template
Only modules with requirements in requirements.json will have sections created
Custom modules (e.g., pm_query, multi_lock) are supported and will be assigned the next available section number

Monorepo Support

For projects with multiple subprojects (monorepo structure):

# Auto-detect monorepo root
python3 scripts/generate.py rtm --project "临床系统" --system "EDC v1.0"

# Explicitly specify monorepo root
python3 scripts/generate.py rtm --project "临床系统" --system "EDC v1.0" \
  --project-root /path/to/monorepo

Detected monorepo layouts:

apps/, packages/, projects/, modules/, services/ directories
Each subproject with its own requirements.json

Template Versioning

Templates are versioned for compatibility tracking:

# Check compatibility (automatic on generate)
python3 scripts/generate.py rtm --project "系统" --system "v1.0" --category 4

# Specify template version
python3 scripts/generate.py rtm --template-version 1.2.0 \
  --project "系统" --system "v1.0" --category 4

Automatic migrations:

Templates are automatically migrated to current version
Version checks prevent incompatible generator/template combinations

CI/CD Integration

Use provided CI/CD templates for automated documentation generation:

GitHub Actions (templates/ci/github-actions.yml):

- uses: actions/checkout@v4
- uses: actions/setup-python@v5
  with:
    python-version: '3.11'
- run: pip install -r requirements.txt
- run: python scripts/generate.py check --output-format json
- run: python scripts/generate.py all --diff-only

GitLab CI (templates/ci/gitlab-ci.yml):

generate:rtm:
  extends: .csv-docs-base
  script:
    - python scripts/generate.py rtm --diff-only
  artifacts:
    paths:
      - validation/RTM*.xlsx

Important Notes

Template Requirements Are Examples

The URS/FS/TS IDs and descriptions in templates are examples only. For actual projects:

Review and modify template requirements based on real system needs
Delete or replace example requirements as appropriate
Use --sync-template-to-db to populate initial requirements, then customize

Change-Driven Risk Assessment (GxP)

Per GxP requirements, any system change requires a risk assessment. When generating RA documents, include the change description:

python3 scripts/generate.py ra \
  --project "临床系统" \
  --system "EDC v1.0" \
  --change-description "升级数据库从 v12 到 v14" \
  --output ./validation/

Parsing Requirements from Code

# Navigate to skill directory
cd <SKILL_DIR>/csv-documentation-generator

# Parse source code for requirements using csv-docs CLI
python3 scripts/cli.py parse ./src

# View extracted requirements
cat requirements.json

Troubleshooting

Common Issues

Issue	Solution
python-docx not installed	Run: `pip install -r requirements.txt`
Encoding errors	Ensure UTF-8 encoding in your terminal
Template not found	Check you are in the correct directory

Getting Help

Run with verbose output:

python3 scripts/generate.py --help

Examples

AI Agent (Non-Interactive Mode)

For AI agents running in non-interactive mode, use --auto-add to automatically add all parsed requirements:

# Parse and auto-add all requirements (non-interactive)
python3 scripts/cli.py parse ./src --auto-add

# Generate with verbose output
python3 scripts/generate.py all --project "XX系统" --system "System v1.0" --category 4 --verbose

Example 1: Generate URS for CTMS

python3 scripts/generate.py urs \
  --project "临床试验管理系统" \
  --system "CTMS v2.0" \
  --category 4 \
  --bilingual true \
  --output ./validation/

Example 2: Generate Risk Assessment

python3 scripts/generate.py ra \
  --project "质量管理系统" \
  --system "QMS v1.5" \
  --category 4 \
  --critical-functions "数据录入,审计追踪,权限控制" \
  --output ./validation/

Example 3: Generate Traceability Matrix

python3 scripts/generate.py rtm \
  --project "实验室系统" \
  --system "LIMS v3.0" \
  --category 4 \
  --urs-file ./URS-001.md \
  --fs-file ./FS-001.md \
  --output ./validation/

About this skill

Automatic

Clinical / Healthcare pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/2/2026

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

LeoYeAI / openclaw-master-skills

skills/csv-documentation-generator/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1Bundled 23 files · scripts

Illustrative Want to see what leoyeai/csv-documentation-generator 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

CSV Documentation Generator

When to Use

Supported Systems

Quick Start

Parse Requirements from Code (pre-generation)

Auto Environment Setup

Interactive GAMP Category Selection

Available Commands

Command Options

Output Format Standards

Document Naming Convention

Output Directory Structure

Document Styling Standards

Template Variables

Regulatory Compliance

GAMP 5 Categories

Key Regulatory Requirements

Semantic Action Graph

Action Graph

Semantic Actions

Data Flow Semantics

Prompt Library

Reference Documents

Example Templates

Bilingual Format

Bilingual Template Format

Requirements Traceability

Automated Traceability Chain

Code Comment Standards (AI Agent Must Follow)

Standard Modules

System Prompt Integration (Manual Configuration Required)

Test Case ID Format

Auto-Sync Feature

Sync Template to Database

Interactive Mode

Compliance Check

Incremental Update (Smart Rebuild)

Git Hooks (Automated Compliance)

Bidirectional Sync

Monorepo Support

Template Versioning

CI/CD Integration

Important Notes

Template Requirements Are Examples

Change-Driven Risk Assessment (GxP)

Parsing Requirements from Code

Troubleshooting

Common Issues

Getting Help

Examples

AI Agent (Non-Interactive Mode)

Example 1: Generate URS for CTMS

Example 2: Generate Risk Assessment

Example 3: Generate Traceability Matrix

SkillSafety

Source

Related skills

More in the registry