Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Universal LaTeX document skill: create, compile, and convert any document to professional PDF with PNG previews. Supports resumes, reports, cover letters, invoices, academic papers, theses/dissertations, academic CVs, presentations (Beamer), scientific posters, formal letters, exams/quizzes, books, cheat sheets, reference cards, exam formula sheets, fillable PDF forms (hyperref form fields), conditional content (etoolbox toggles), mail merge from CSV/JSON (Jinja2 templates), version diffing (lat
SkillScore77/100
Test results 77
3/3 safety checks passed
27%77%of test cases passed +50 pts
Passed 17 of 22 test cases with the skill, and 6 of 22 without it.
11 fixed6 passed either way5 still failing0 broke
One test taskI need to design an A0 portrait poster for an AI conference using the poster sub-workflow. What interactive step must be performed before customizing template content?
Adds every file the skill has, pinned so you can update it. No account needed.
pip install "decimalai>=0.15"
decimalai skills add ndpvt-web-latex-document-skill
claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"The same install from Python, for a notebook or a setup script.
pip install "decimalai>=0.15"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
if result.exit_code:
raise SystemExit(result.problems)
print("Added. Start Claude Code with the command below.")claude --plugin-dir "$(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.
/plugin marketplace add decimal-labs/decimalai-claude-plugins
/plugin install decimalai@decimalai/decimalai:install ndpvt-web-latex-document-skillThe plugin needs python3 and curl on your machine.
No Python or account. Claude Code picks it up in its next session.
curl -fsSL --create-dirs -o .claude/skills/ndpvt-web-latex-document-skill/SKILL.md https://app.decimal.ai/s/ndpvt-web-latex-document-skill/SKILL.md.claude/skills/ndpvt-web-latex-document-skill/SKILL.mdThis gets SKILL.md only. The skill's 30 other files are not included.
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.
pip install "decimalai[claude-agent-sdk]"
decimalai skills add ndpvt-web-latex-document-skillimport 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.
pip install "decimalai[claude-agent-sdk]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdimport asyncio
from pathlib import Path
from claude_agent_sdk import ClaudeAgentOptions, query
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
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.
pip install "decimalai[openai-agents]"
decimalai skills add ndpvt-web-latex-document-skillfrom 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.
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.
pip install "decimalai[openai-agents]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from agents import Agent, Runner
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
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.
pip install "decimalai[openai]"
decimalai skills add ndpvt-web-latex-document-skillimport 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.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.
pip install "decimalai[openai]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from openai import OpenAI
client = OpenAI()
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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].messageThis gets SKILL.md only. The skill's 30 other files are not included.
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.
pip install "decimalai[anthropic]"
decimalai skills add ndpvt-web-latex-document-skillimport 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()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.
pip install "decimalai[anthropic]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
import anthropic
client = anthropic.Anthropic()
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
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.
pip install "decimalai[google-genai]"
decimalai skills add ndpvt-web-latex-document-skillfrom 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)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.
pip install "decimalai[google-genai]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from google import genai
from google.genai import types
client = genai.Client()
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
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.
pip install "decimalai[langchain]" langchain-openai
decimalai skills add ndpvt-web-latex-document-skillfrom 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!"}]})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.
pip install "decimalai[langchain]" langchain-openaifrom decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from langchain.agents import create_agent
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.
pip install "decimalai>=0.16" "crewai>=1.15.27"
decimalai skills add ndpvt-web-latex-document-skillimport 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.
pip install "decimalai>=0.16" "crewai>=1.15.27"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom 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 30 other files are not included.
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.
pip install "decimalai[adk]"
decimalai skills add ndpvt-web-latex-document-skillfrom 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.
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.
pip install "decimalai[adk]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from google.adk.agents import LlmAgent
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
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.
pip install "decimalai[pydantic-ai]"
decimalai skills add ndpvt-web-latex-document-skillfrom 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.
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.
pip install "decimalai[pydantic-ai]"from decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from pydantic_ai import Agent
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.
pip install "decimalai>=0.16" deepagents
decimalai skills add ndpvt-web-latex-document-skillimport 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.
pip install "decimalai>=0.16" deepagentsfrom decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from deepagents import create_deep_agent
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
Its own skill loader reads the installed folder. No DecimalAI account needed.
pip install "decimalai>=0.16" agent-framework
decimalai skills add ndpvt-web-latex-document-skillimport 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.
pip install "decimalai>=0.16" agent-frameworkfrom decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdimport asyncio
from pathlib import Path
from agent_framework import Agent
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
Strands' own skill loader reads the installed folder. No DecimalAI account needed.
pip install "decimalai>=0.16" strands-agents
decimalai skills add ndpvt-web-latex-document-skillfrom 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.
pip install "decimalai>=0.16" strands-agentsfrom decimalai import skillpkg
# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("ndpvt-web-latex-document-skill")
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.
.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdfrom pathlib import Path
from strands import Agent
skill = Path(".agents/skills/ndpvt-web-latex-document-skill/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 30 other files are not included.
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 ndpvt-web-latex-document-skillNo Python or account. Cursor, Copilot, Cline and Warp read this folder by themselves.
curl -fsSL --create-dirs -o .agents/skills/ndpvt-web-latex-document-skill/SKILL.md https://app.decimal.ai/s/ndpvt-web-latex-document-skill/SKILL.md.agents/skills/ndpvt-web-latex-document-skill/SKILL.mdThis gets SKILL.md only. The skill's 30 other files are not included.
Give its text to your model as instructions. No account or package needed.
This gets SKILL.md only. The skill's 30 other files are not included.
Raw files: /s/ndpvt-web-latex-document-skill/SKILL.md · /s/ndpvt-web-latex-document-skill.json
Fetch the file and give its text to your model as instructions. No account or package needed.
curl -fsSL https://app.decimal.ai/s/ndpvt-web-latex-document-skill/SKILL.mdcurl -fsSL https://app.decimal.ai/s/ndpvt-web-latex-document-skill.jsonThis gets SKILL.md only. The skill's 30 other files are not included.
You don't need a DecimalAI account.
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.
Create any LaTeX document, compile to PDF, and generate PNG previews. Convert PDFs of any type to LaTeX.
examples/ieee-twocolumn-sample.tex for a working IEEEtran baseline and use examples/ieee-twocolumn.png / examples/ieee-twocolumn-p1.png / examples/ieee-twocolumn-p2.png as visual references.assets/templates/ or write from scratchpython3 <skill_path>/../generate-image/scripts/generate_image.py "prompt" --output ./outputs/figure.pngpython3 <skill_path>/scripts/generate_chart.py <type> --data '<json>' --output chart.pngbash <skill_path>/scripts/mermaid_to_image.sh diagram.mmd output.png</> in text mode, vary section formats, limit \newpage, size images at 0.75-0.85 textwidth.scripts/compile_latex.sh (auto-detects XeLaTeX for CJK/RTL, glossaries, bibliography)When the user requests a poster: read references/poster-design-guide.md for the complete workflow including conference size presets (NeurIPS/ICML/CVPR/ICLR dimensions), layout archetypes (Traditional/BetterPoster/Visual-Heavy), color schemes, and typography standards. Use poster.tex (portrait) or poster-landscape.tex (landscape). Ask the user for conference/orientation, layout style, and color scheme using AskUserQuestion, then proceed to step 5.
When the user requests a cheat sheet, reference card, or formula sheet:
cheatsheet.tex (general, 3-col landscape), cheatsheet-exam.tex (exam formula, 2-col portrait), or cheatsheet-code.tex (programming, 4-col landscape).Generate N personalized documents from a LaTeX template + CSV/JSON data source using scripts/mail_merge.py. Template syntax: {{variable}} for simple substitution, Jinja2 (<< >>, <% %>) for conditionals/loops. See assets/templates/mail-merge-letter.tex for an example. Full guide: references/interactive-features.md.
Generate highlighted change-tracked PDFs using scripts/latex_diff.sh. Supports file-to-file diff, git revision diff, multi-file flatten, and custom markup styles. Full guide: references/interactive-features.md.
Convert between Markdown, DOCX, HTML, and LaTeX using scripts/convert_document.sh. Full guide: references/format-conversion.md.
Convert existing PDFs (handwritten notes, printed reports, legal docs) to LaTeX. Full pipeline: references/pdf-conversion.md.
Quick steps: Split PDF into page images (scripts/pdf_to_images.sh), select a conversion profile, create shared preamble, apply scaling strategy, validate with scripts/validate_latex.py, concatenate, compile.
Scaling strategy: 1-10 pages = single agent; 11-20 pages = split in half (2 agents); 21+ pages = batch-7 pipeline (ceil(N/7) agents with run_in_background: true).
Conversion profiles (in references/profiles/): math-notes.md (equations, theorems -- has beautiful mode), business-document.md (reports, memos), legal-document.md (contracts, statutes), general-notes.md (handwritten, mixed content).
Fill existing PDF forms -- both fillable (with form fields) and non-fillable (image-based). Full guide: references/pdf-operations.md.
Step 1: Check form type:
python3 <skill_path>/scripts/pdf_check_form.py form.pdf
If fillable (has form fields):
# Extract field metadata
python3 <skill_path>/scripts/pdf_extract_fields.py form.pdf field_info.json
# Create field_values.json with values for each field, then fill
python3 <skill_path>/scripts/pdf_fill_form.py form.pdf field_values.json output.pdf
If non-fillable (no form fields):
# Convert to images for visual analysis
bash <skill_path>/scripts/pdf_to_images.sh form.pdf ./tmp/pages
# Visually identify fields, create fields.json with bounding boxes
# Validate bounding boxes (+ optional validation image)
python3 <skill_path>/scripts/pdf_validate_boxes.py fields.json --image page_1.png --output validation.png --page 1
# Fill with text annotations
python3 <skill_path>/scripts/pdf_fill_annotations.py form.pdf fields.json output.pdf
For text/table extraction (pdfplumber), OCR (pytesseract), programmatic PDF creation (reportlab), watermarking, page rotation/cropping, metadata extraction, JavaScript libraries (pdf-lib, pdfjs-dist), and batch processing, see references/pdf-operations.md.
# Basic compile (auto-detects engine)
bash <skill_path>/scripts/compile_latex.sh document.tex
# Compile + generate PNG previews
bash <skill_path>/scripts/compile_latex.sh document.tex --preview
# Compile + PNG in specific directory
bash <skill_path>/scripts/compile_latex.sh document.tex --preview --preview-dir ./outputs
# Force a specific engine
bash <skill_path>/scripts/compile_latex.sh document.tex --engine xelatex
bash <skill_path>/scripts/compile_latex.sh document.tex --engine lualatex
# Use latexmk for automatic multi-pass (recommended for complex documents)
bash <skill_path>/scripts/compile_latex.sh document.tex --use-latexmk --preview
# PDF/A output for thesis submissions and archival compliance
bash <skill_path>/scripts/compile_latex.sh document.tex --pdfa
# Verbose output for debugging compilation issues
bash <skill_path>/scripts/compile_latex.sh document.tex --verbose
# Quiet mode for batch/CI jobs (only errors shown)
bash <skill_path>/scripts/compile_latex.sh document.tex --quiet
# Clean auxiliary files only (no compilation)
bash <skill_path>/scripts/compile_latex.sh document.tex --clean
| Flag | Description |
|---|---|
--preview | Generate PNG previews of each page after compilation |
--preview-dir DIR | Directory for PNG output (default: same as .tex file) |
--engine ENGINE | Force engine: pdflatex, xelatex, or lualatex |
--use-latexmk | Use latexmk as compilation backend (auto multi-pass, bibliography, index) |
--verbose | Show full compilation output (all engine logs) |
--quiet | Suppress all non-error output |
--clean | Remove auxiliary files (.aux, .log, .bbl, .fdb_latexmk, etc.) and exit |
--pdfa | Produce PDF/A-2b compliant output (auto-injects pdfx package) |
--auto-fix | Auto-fix common compilation errors (float placement, encoding) |
Manual multi-pass (default): Runs the engine multiple times with bibliography/index/glossary passes as needed. This is the traditional approach and works without latexmk installed.
latexmk (--use-latexmk): Uses latexmk for automatic dependency-driven compilation. Recommended for complex documents with bibliographies, indexes, glossaries, or cross-references -- latexmk determines the correct number of passes automatically. Requires latexmk (included with TeX Live).
When texfot is installed (included with TeX Live), compilation output is automatically filtered to show only relevant warnings and errors, suppressing noisy package loading messages. This applies in the default verbosity mode. Use --verbose to see unfiltered output, or --quiet to suppress all non-error output.
Engine auto-detection: If the .tex file uses fontspec, xeCJK, or polyglossia, the script automatically uses xelatex. If it uses luacode or luatextra, it uses lualatex. Otherwise defaults to pdflatex. Override with --engine.
The script auto-installs texlive (including texlive-science, texlive-xetex, texlive-luatex, biber) and poppler-utils if missing. It auto-detects \bibliography{} (runs bibtex), \addbibresource{} (runs biber), \makeindex (runs makeindex), \makeglossaries (runs makeglossaries), runs the correct number of passes, generates PNG previews with pdftoppm, and cleans auxiliary files.
For PDF utilities (encrypt, merge, optimize, extract pages, pdf-to-images), LaTeX quality tools (lint, word count, analysis, package check, citations), compilation auto-fix, bibliography fetching, and diagram scripts (Mermaid, Graphviz, PlantUML), see references/script-tools.md.
Copy from assets/templates/ and customize.
Select based on experience level, industry, and ATS requirements. See references/resume-ats-guide.md for full ATS guidance.
| Template | Best For | Key Feature | ATS Score |
|---|---|---|---|
resume-classic-ats.tex | Finance, law, government, any ATS portal | Zero graphics, plain text only, maximum parse safety | 10/10 |
resume-modern-professional.tex | Tech, corporate, general professional | Subtle color accents, clean design, good ATS + human appeal | 9/10 |
resume-executive.tex | VP, Director, C-suite (5-15+ years) | Two-page, executive summary, board roles, P&L focus | 9/10 |
resume-technical.tex | Software, data, engineering roles | Skills-first hybrid, projects section, tech stack emphasis | 9/10 |
resume-entry-level.tex | New graduates, career starters | Education-first, one page, coursework, activities | 9/10 |
All 5 templates follow ATS rules: single-column, no graphics/images, no tables for layout, standard section headings, contact info in body (not header/footer).
homework.tex -- Homework/assignment submission template (article class, 11pt) with toggle-able solutions (\showsolutionstrue/\showsolutionsfalse), custom problem/solution environments, honor code section, fancyhdr headers, amsmath+amssymb+amsthm math, listings code highlighting (Python, Java, C++, Matlab styles), enumitem for (a), (b), (c) sub-parts, hyperref. Customization via \coursename, \assignmentnumber, \studentname, \studentid, \duedate. Example problems: calculus, proof by induction, Python coding, matrix algebra, kinematics, Java OOP.lab-report.tex -- STEM lab report template (article class, 11pt) with siunitx for uncertainties and SI units, pgfplots for data visualization, booktabs for professional tables, amsmath for equations, fancyhdr headers, caption+subcaption for figures. Structured sections: abstract, theory, procedure, data/results (with uncertainty tables), analysis (pgfplots graphs), discussion (error analysis), conclusion, references, appendix. Example: pendulum period measurement with error propagation.lecture-notes.tex -- Beautiful lecture notes (scrartcl KOMA-Script class) with Palatino font, microtype typography, color-coded tcolorbox theorem environments (blue theorems, green definitions, orange examples, purple remarks), TikZ graph theory diagram styles and macros (\CompleteGraph, \CycleGraph, \PathGraph), styled section headings with titlesec, fancyhdr headers, hyperref + cleveref navigation. Ideal for converting handwritten math/science notes to beautiful PDFs. Includes graph theory custom commands (\V, \E, \deg, \chr). Used by the math-notes conversion profile in beautiful mode.thesis.tex -- Thesis/dissertation (book class) with title page, declaration, abstract, acknowledgments, TOC, list of figures/tables, chapters, appendices, bibliography. Front matter uses roman numerals, main matter uses arabic. Includes theorem environments.academic-cv.tex -- Multi-page academic CV with publications (journal/conference/preprint sections), grants and funding, teaching, advising (current/graduated students), awards, professional service, invited talks. ORCID and Google Scholar links.book.tex -- Full book (book class) with half-title, title page, copyright page, dedication, preface, acknowledgments, TOC, list of figures/tables, parts, chapters, appendix, bibliography, index. Custom chapter headings, epigraphs, fancyhdr, microtype.poster.tex -- Conference poster (tikzposter class, A0 portrait) with 2-column layout, QR code, 5 color schemes, tikzfigure charts, tables, coloredbox highlights. Includes commented #BetterPoster layout variant (central billboard + sidebars). Portrait is standard for most conferences. See poster design guide for conference size presets.poster-landscape.tex -- Landscape conference poster (tikzposter class, A0 landscape) with 3-column (30/40/30) layout, QR code, tech purple color scheme. For CS/ML conferences (NeurIPS, ICML, CVPR, ICLR). Includes commented \geometry{} presets for CVPR/NeurIPS custom sizes.cheatsheet.tex -- General reference card (extarticle class, landscape A4, 7pt base font) with 3-column multicols layout, tcolorbox colored section boxes, tight spacing. For command references, concept summaries, quick-reference guides. Supports front+back printing.cheatsheet-exam.tex -- Exam formula sheet (extarticle class, portrait A4, 6pt base font) with 2-column layout, maximum content density, black-and-white printer-safe. For university exam "allowed cheat sheets" with formulas, theorems, definitions. Optimized for single-sided printing.cheatsheet-code.tex -- Programming reference card (extarticle class, landscape A4, 7pt base font) with 4-column layout, minted syntax highlighting, monospace font emphasis. For language syntax, API reference, CLI commands, keyboard shortcuts.letter.tex -- Formal business letter with colored letterhead bar, TikZ logo placeholder, company info, recipient block, date, subject line, signature. Professional corporate appearance.exam.tex -- Exam/quiz (exam class) with grading table, multiple question types (multiple choice, true/false, fill-in-blank, matching, short answer, essay), point values, solution toggle via \printanswers.report.tex -- Business report with TOC, headers/footers, data tables, bar chart (pgfplots), process flowchart (TikZ), recommendationscover-letter.tex -- Professional cover letter with sender/recipient blocksinvoice.tex -- Invoice with company header, line items table, subtotal/tax/totalacademic-paper.tex -- Research paper with abstract, sections, bibliography, figures. Includes example .bib file (references.bib) for BibTeX citations.examples/ieee-twocolumn-sample.tex -- IEEE Transactions / journal example using \documentclass[journal,10pt]{IEEEtran} with abstract, keywords, single-column figure, single-column table, full-width table, and numeric bibliography. Use with references/ieee-journal-twocolumn-guide.md. Preview PNGs: examples/ieee-twocolumn.png, examples/ieee-twocolumn-p1.png, examples/ieee-twocolumn-p2.png.presentation.tex -- Beamer presentation with title slide, content frames, columnsfillable-form.tex -- Fillable PDF form (article class) with hyperref form fields: text inputs, checkboxes, radio buttons, dropdowns, push buttons. Two-column layout with tabularx. Custom \FormField, \FormCheck, \FormDropdown helper commands. Requires Adobe Reader for full form support (browser PDF viewers have limited form capabilities).conditional-document.tex -- Configurable document (article class) with etoolbox toggle system: show/hide TOC, appendix, watermark, draft mode, confidential marking, abstract. Template variables via \providecommand (title, author, org, version). Three visual profiles (corporate, academic, minimal). Supports command-line variable passing for CI/CD.mail-merge-letter.tex -- Mail merge letter template with {{variable}} placeholders for use with scripts/mail_merge.py. Company-branded letterhead, recipient address block, body text with data-driven personalization. Pair with CSV/JSON data source.resume.tex -- Legacy resume template (has photo area and tables -- not ATS-optimized)Usage:
cp <skill_path>/assets/templates/resume-classic-ats.tex ./outputs/my_resume.tex
# Edit content, then compile
bash <skill_path>/scripts/compile_latex.sh ./outputs/my_resume.tex --preview --preview-dir ./outputs
| User Request | Template | Document Class |
|---|---|---|
| Resume, CV (ATS-safe) | resume-classic-ats.tex | article |
| Resume (modern look) | resume-modern-professional.tex | article |
| Resume (senior/executive) | resume-executive.tex | article |
| Resume (technical/engineering) | resume-technical.tex | article |
| Resume (new graduate) | resume-entry-level.tex | article |
| Homework, assignment, problem set | homework.tex | article |
| Lab report, experiment writeup | lab-report.tex | article |
| Lecture notes, math notes (beautiful) | lecture-notes.tex | scrartcl |
| Thesis, dissertation | thesis.tex | book |
| Academic CV (publications, grants) | academic-cv.tex | article |
| Report, analysis | report.tex | article |
| Cover letter | cover-letter.tex | article |
| Invoice | invoice.tex | article |
| Academic paper | academic-paper.tex + references.bib | article |
| IEEE journal, IEEE Transactions, IEEE two-column paper | examples/ieee-twocolumn-sample.tex + ieee-journal-twocolumn-guide.md | IEEEtran |
| Book | book.tex | book |
| Scientific poster (portrait) | poster.tex | tikzposter |
| Scientific poster (landscape) | poster-landscape.tex | tikzposter |
| Cheat sheet, reference card | cheatsheet.tex | extarticle |
| Exam formula sheet, crib sheet | cheatsheet-exam.tex | extarticle |
| Programming reference card | cheatsheet-code.tex | extarticle |
| Formal business letter | letter.tex | article |
| Exam, quiz, test | exam.tex | exam |
| Slides, presentation | presentation.tex | beamer |
| Fillable PDF form | fillable-form.tex | article |
| Conditional/configurable document | conditional-document.tex | article |
| Mail merge, batch letters | mail-merge-letter.tex + scripts/mail_merge.py | article |
| Version diff (latexdiff) | Use scripts/latex_diff.sh + see interactive-features.md | varies |
| Convert PDF to LaTeX | Select profile + see pdf-conversion.md | varies |
| Convert formats | Use scripts/convert_document.sh + see format-conversion.md | varies |
| Mail merge from CSV/JSON | Use scripts/mail_merge.py + see interactive-features.md | varies |
Always escape: % → \%, $ → \$, & → \&, # → \#, _ → \_
Angle brackets in text mode: < and > are NOT valid in LaTeX text mode with T1 encoding. They render as inverted question marks (¡ or ¿). Always use math mode or text commands:
<5% → $<$5\% or \textless 5\%>50 → $>$50 or \textgreater 50<$300 → $<$\$300>= → $\geq$, <= → $\leq$This is one of the most common silent errors in generated LaTeX — the document compiles without errors but the PDF shows garbage characters.
Date ranges use en-dash: 2019--2025 (double hyphen).
\documentclass[11pt,a4paper]{article}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{geometry}
\usepackage{hyperref}
\usepackage{xcolor}
\usepackage{graphicx}
\usepackage{tabularx}
\usepackage{colortbl}
\usepackage{enumitem}
\usepackage{titlesec}
For ready-to-use LaTeX code snippets (tables, charts, flowcharts, bibliography, watermarks, landscape pages, multi-language, images, Mermaid diagrams, matplotlib charts, CSV tables, AI images, fillable forms, conditional content, mail merge, version diff), read references/code-patterns.md.
When creating reports, use the enrichment elements the user selected in step 4. For the quick reference table of visual element tools (charts, flowcharts, AI images, tables, timelines) and sizing guidelines, see references/advanced-features.md. Use \begin{figure}[H] (from float package) to prevent figures from floating. Size TikZ diagrams with width=0.8\textwidth or smaller.
When uncertain about LaTeX patterns, compilation issues, or document quality, consult this table and load relevant files.
| Topic | File | When to Read |
|---|---|---|
| Long-Form Best Practices | long-form-best-practices.md | MUST READ for 5+ page documents: 9 anti-patterns, report preamble, content generation checklist |
| Code Patterns (LaTeX snippets) | code-patterns.md | Ready-to-use LaTeX code: tables, charts, flowcharts, bibliography, watermarks, landscape, multi-lang, images, forms, mail merge, diffs |
| Script & Tool Reference | script-tools.md | PDF utilities (encrypt, merge, optimize, extract), LaTeX quality tools (lint, wordcount, analysis), compilation auto-fix, bibliography, diagrams |
| Bibliography/Citations | bibliography-guide.md | BibTeX/biblatex, citation styles, .bib format |
| Advanced Features & Visual Elements | advanced-features.md | Watermarks, landscape, multi-lang, code, algorithms, tcolorbox, siunitx, advanced charts, AI images, visual elements reference table |
| Mermaid Diagrams | mermaid-diagrams.md | Flowcharts, sequence, class, ER, Gantt, pie, mindmap |
| Python Charts (matplotlib) | python-charts.md | Bar, line, scatter, pie, heatmap, box, histogram, area, radar |
| Format Conversion (Pandoc) | format-conversion.md | Markdown/DOCX/HTML to/from LaTeX |
| Tables and Images | tables-and-images.md | Colored rows, multi-row/column, booktabs, long tables, images, TikZ |
| Charts and Graphs (pgfplots) | charts-and-graphs.md | Line plots, bar charts, scatter plots, pie charts in pgfplots |
| IEEE Journal Two-Column Typography | ieee-journal-twocolumn-guide.md | IEEEtran journal layout, float rules, XeLaTeX vs. pdfLaTeX, bibliography, Markdown-to-IEEE workflow, sample PNGs in examples/ieee-twocolumn*.png |
| LaTeX Packages | packages.md | Common packages reference |
| Poster Design Guide | poster-design-guide.md | Conference size presets, typography, color schemes, layout archetypes, content guidelines |
| Resume ATS Guide | resume-ats-guide.md | ATS rules, LaTeX pitfalls, keywords |
| PDF-to-LaTeX Pipeline | pdf-conversion.md | Full conversion pipeline, scaling strategy, worker agents, validation |
| Interactive Features | interactive-features.md | Fillable PDF forms, conditional content, mail merge from CSV/JSON, latexdiff version tracking |
| Cheat Sheet Design | cheatsheet-guide.md | Template selection, typography, density optimization, PDF-to-cheatsheet pipeline |
| PDF-to-Cheatsheet Prompts | pdf-extraction-prompts.md | LLM prompts for extraction stages: structure analysis, subject-specific extraction, compression, formatting |
| Visual Packages (TikZ) | visual-packages.md | 24 installed TikZ/visualization packages with working examples |
| Graphviz & PlantUML | graphviz-plantuml.md | .dot to PDF/PNG, .puml to PDF/PNG, LaTeX integration |
| Debugging & Errors | debugging-guide.md | Reading LaTeX errors, 20 common errors, .log analysis, silent failures |
| Accessibility & PDF/A | accessibility-guide.md | PDF/A for thesis, PDF/UA for screen readers, tagged PDFs, alt text |
| Beamer Presentations | beamer-guide.md | Themes, overlays/animations, code slides, handouts, presenter notes |
| Font Selection | font-guide.md | Font families, fontspec, mathematical fonts, fontawesome5, microtype, CJK |
| Collaboration & CI/CD | collaboration-guide.md | Git for LaTeX, GitHub Actions, Docker, multi-author workflows |
| PDF Operations | pdf-operations.md | Form filling (fillable + non-fillable), text/table extraction (pdfplumber), OCR (pytesseract), programmatic PDF creation (reportlab), watermarking, rotation/cropping, metadata, JavaScript (pdf-lib, pdfjs-dist), batch processing, performance tips |
./outputs/ for user visibilitypoppler-utils (auto-installed by script)These cause Undefined control sequence -- always include the required package:
| If you use... | You MUST include | Error without it |
|---|---|---|
\rowcolor{} | \usepackage{colortbl} | Undefined control sequence \rowcolor |
\url{} in .bib with natbib | \usepackage{url} | Undefined control sequence \url |
\checkmark | \usepackage{amssymb} | Undefined control sequence \checkmark |
\begin{figure}[H] | \usepackage{float} | Unknown float option H |
\rowcolors{}{}{} | \usepackage[table]{xcolor} or \usepackage{colortbl} | Undefined control sequence |
hyperref is fine for normal documents (most templates use it).\set@color errors.-interaction=nonstopmode (not -halt-on-error) to ensure PDFs are produced even with warnings. This is intentional -- many documents produce warnings on first pass that resolve on subsequent passes.setup.sh) verifies their availability. latexmk is used when --use-latexmk is passed; texfot is used automatically when available (default verbosity mode)..chktexrc configuration file is included at the skill root to suppress common false positives when linting generated LaTeX documents. The latex_lint.sh script uses it automatically.For documents 5+ pages (reports, theses, books), read references/long-form-best-practices.md for 9 critical anti-patterns, the report preamble best practices, and the content generation checklist. This is MUST-READ material before generating any long document.
Published skills whose descriptions closely match this one, with their headline benchmark lift.
Skills that do similar work, with their headline benchmark lift.
From the same source and nearby in the catalogue.