Ready to supercharge your AI agents?

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

Get Started Free →

Skill: Calibrate from RTSP Streams · nvidia | DecimalAI

Scienceby nvidia

Skill: Calibrate from RTSP Streams

Calibrate a new dataset from live RTSP camera streams via the AutoMagicCalib REST API.

Never made a case worseMaintained

SkillScore

SkillScore68/100

Test results 68

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

14%68%of test cases passed +55 pts

Without skill3 of 22

With skill15 of 22

12 fixed3 passed either way7 still failing0 broke

Measured on gemini-3.6-flash9/8/2026v118 comparable

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskWe ingested RTSP stream clips into project 'plaza_south' and uploaded `alignment_data.json` and `layout.png`. No `settings.json` file was uploaded. We are ready to call `/v1/calibrate/plaza_south`. What parameter must be resolved before initiating calibration?

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 nvidia-amc-run-rtsp-calibration
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("nvidia-amc-run-rtsp-calibration")
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 nvidia-amc-run-rtsp-calibration

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/nvidia-amc-run-rtsp-calibration/SKILL.md https://app.decimal.ai/s/nvidia-amc-run-rtsp-calibration/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/nvidia-amc-run-rtsp-calibration/SKILL.md

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

Other ways to add it

Claude Code docs

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

Install

pip install "decimalai[claude-agent-sdk]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/nvidia-amc-run-rtsp-calibration/SKILL.md").read_text()
options = ClaudeAgentOptions(system_prompt=f"You are a helpful agent.\n\n{skill}", max_turns=6)

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

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

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

Other ways to add it

Claude Agent SDK docs

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

Install

pip install "decimalai[openai-agents]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

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

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

Other ways to add it

OpenAI Agents docs

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

Install

pip install "decimalai[openai]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

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

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

Other ways to add it

OpenAI SDK docs

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

Install

pip install "decimalai[anthropic]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

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

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

Other ways to add it

Anthropic docs

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

Install

pip install "decimalai[google-genai]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/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/nvidia-amc-run-rtsp-calibration/SKILL.md").read_text()
response = client.models.generate_content(
    model="gemini-3.5-flash",
    contents="Hello!",
    config=types.GenerateContentConfig(system_instruction=f"You are a helpful agent.\n\n{skill}"),
)

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

Other ways to add it

Gemini API docs

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

Install

pip install "decimalai[langchain]" langchain-openai
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

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

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

Other ways to add it

LangChain docs

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

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

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

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

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

Other ways to add it

CrewAI docs

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

Install

pip install "decimalai[adk]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/nvidia-amc-run-rtsp-calibration/SKILL.md").read_text()
agent = LlmAgent(
    name="support",
    model="gemini-3.5-flash",
    instruction=lambda context: f"You are a helpful agent.\n\n{skill}",
)

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

Other ways to add it

Google ADK docs

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

Install

pip install "decimalai[pydantic-ai]"
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

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

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

Other ways to add it

Pydantic AI docs

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

Install

pip install "decimalai>=0.16" deepagents
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/nvidia-amc-run-rtsp-calibration/SKILL.md").read_text()
agent = create_deep_agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

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

Other ways to add it

Deep Agents docs

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

Install

pip install "decimalai>=0.16" agent-framework
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/nvidia-amc-run-rtsp-calibration/SKILL.md").read_text()
agent = Agent(client=client, instructions=f"You are a helpful agent.\n\n{skill}")
asyncio.run(agent.run("Hello!"))

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

Other ways to add it

Microsoft Agent Framework docs

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

Install

pip install "decimalai>=0.16" strands-agents
decimalai skills add nvidia-amc-run-rtsp-calibration

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("nvidia-amc-run-rtsp-calibration")
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/nvidia-amc-run-rtsp-calibration/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/nvidia-amc-run-rtsp-calibration/SKILL.md").read_text()
agent = Agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent("Hello!")

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

Other ways to add it

Strands docs

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

pip install "decimalai>=0.15"
decimalai skills add nvidia-amc-run-rtsp-calibration

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/nvidia-amc-run-rtsp-calibration/SKILL.md https://app.decimal.ai/s/nvidia-amc-run-rtsp-calibration/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/nvidia-amc-run-rtsp-calibration/SKILL.md

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

Other ways to add it

Cursor and other editors docs

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

Get the file

curl -fsSL https://app.decimal.ai/s/nvidia-amc-run-rtsp-calibration/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/nvidia-amc-run-rtsp-calibration.json

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

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 9/8/2026
Source: nvidia/skills ↗
License: Apache-2.0

Manage it in DecimalAI

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

Use this skill

SKILL.md

Skill: Calibrate from RTSP Streams

When to Use This Skill

Activate this skill when the user wants to calibrate from live RTSP camera streams. Typical prompts:

"calibrate RTSP streams" / "calibrate from live cameras"
"run AMC on RTSP"
The user provides one or more rtsp://... URLs

VIOS records fixed-duration clips from each stream, the AMC microservice ingests those clips into a project, then the workflow follows the same verification, calibration, polling, and results path as pre-recorded MP4 calibration.

Do not use this skill for local MP4 files already on disk; route those requests to skills/amc-run-video-calibration/SKILL.md. Do not use it for the bundled sample dataset; route that to skills/amc-run-sample-calibration/SKILL.md.

Never reuse files from the bundled sample dataset, extracted sample zip, assets/, or previous projects for RTSP calibration unless the user explicitly provides those paths for this RTSP scene. Similar camera names, stream counts, or cam_00/cam_01 ordering are not evidence that sample alignment, layout, GT, or detector settings apply.

Prerequisites

AMC microservice and UI running (follow skills/amc-setup-calibration-stack/SKILL.md if needed).
VIOS is running and reachable from the AMC microservice.
VIOS_BASE_URL is configured in the AMC microservice environment before capture starts.
RTSP URLs are reachable from the VIOS host.
Camera streams have enough moving people/objects for calibration; record at least 2-3 minutes when possible.
Python 3 with requests installed when using the bundled script.

Data Privacy

RTSP URLs may contain usernames, passwords, hostnames, or network topology. Do not print full RTSP URLs if credentials are embedded. Pass VIOS tokens through environment variables or secure host prompts; do not echo tokens in chat, logs, or final answers.

What to Ask the User

Required

RTSP URLs, one per camera.
Camera names, one per stream. Use cam_00, cam_01, ... if the user does not provide names.
Recording duration in seconds. Minimum is 60; prefer 120-180 or more when the scene has sparse motion.
Microservice URL, for example http://<HOST_IP>:8000 or http://<HOST_IP>:8000/v1.
Project name.
Calibration asset source for this RTSP scene:
- a local directory to scan, such as /data/my_rtsp_calib/;
- explicit paths to settings, alignment, layout, and optional GT files; or
- confirmation that the user will upload/tune settings and alignment in the AMC UI.

If the user does not provide a local asset source, stop and ask whether they want to provide a path or use UI upload. Give the UI link as http://<HOST_IP>:<AUTO_MAGIC_CALIB_UI_PORT>; the default UI port is 5000.

Auto-Detected or Asked

RTSP clips are recorded by VIOS, so there is no local videos directory to anchor file discovery. Only scan a directory the user explicitly provided for this RTSP scene. If the user provides a settings file path, use that file's directory as the scan directory. If the user provides a calibration asset directory, scan only that directory. Otherwise ask this question before planning uploads or calibration:

Do you have a local calibration asset directory or settings file for these RTSP streams, or should you upload/tune settings and alignment in the AMC UI at http://<HOST_IP>:<AUTO_MAGIC_CALIB_UI_PORT>?

File	Candidate filenames	UI fallback
Calibration settings	Explicit user path, or `settings.json`, `config.json`, or `calibration_config.json` in the user-provided asset directory	UI Step 3: Parameters
Alignment JSON	Explicit user path, or `alignment_data.json` in the user-provided asset directory/settings directory	UI Step 4: Alignment
Layout PNG	Explicit user path, or `layout.png` in the user-provided asset directory/settings directory	UI Step 4: Alignment
Ground truth zip	Optional explicit user path, or `GT.zip`/`gt.zip` in the user-provided asset directory	Omit metrics

Posting the settings file replaces UI Step 3 and may pin detector or detector_type. If it pins resnet or transformer, pass that same detector to /calibrate. If no settings file pins a detector, ask the user which detector to use; do not silently default to resnet.

Optional

sensor_id per stream if the cameras are already registered in VIOS. Leave unset for auto-registration.
Ground truth zip (GT.zip) for evaluation metrics.
Focal lengths, one per camera.
VIOS bearer token, if the VIOS deployment requires one.
Whether to run VGGT refinement after AMC completes, only when the project reports vggt_state == "READY".

Instructions

The bundled script in scripts/run_rtsp_calibration.py implements this sequence end to end. Use the prose below for decisions, UI fallback, and troubleshooting.

Step 0 - Verify AMC and VIOS

Confirm the AMC microservice is reachable:

curl -sf http://<HOST_IP>:<MS_PORT>/v1/ready

Confirm VIOS is reachable before starting capture. Probe in this order and stop at the first working URL:

: "${REPO_ROOT:?set REPO_ROOT to the auto-magic-calib checkout. Run amc-setup-calibration-stack Step 0b first.}"
grep -q "AutoMagicCalib" "$REPO_ROOT/README.md" 2>/dev/null && grep -q "auto-magic-calib-ms" "$REPO_ROOT/compose/ms/compose.yml" 2>/dev/null || { echo "ERROR: REPO_ROOT is not an auto-magic-calib checkout: $REPO_ROOT" >&2; exit 1; }
VIOS_BASE_URL=""

# Default local VIOS port.
if curl -sf http://localhost:30888/vst/api/v1/sensor/list >/dev/null 2>&1; then
  HOST_IP=$(grep ^HOST_IP "$REPO_ROOT/compose/.env" 2>/dev/null | cut -d= -f2)
  VIOS_BASE_URL="http://${HOST_IP:-localhost}:30888"
  echo "VIOS detected at $VIOS_BASE_URL"
fi

# Running AMC microservice container environment.
if [ -z "$VIOS_BASE_URL" ]; then
  VIOS_BASE_URL=$(docker exec auto-magic-calib-ms-1 printenv VIOS_BASE_URL 2>/dev/null)
fi

# Compose environment file.
if [ -z "$VIOS_BASE_URL" ]; then
  VIOS_BASE_URL=$(grep ^VIOS_BASE_URL "$REPO_ROOT/compose/.env" 2>/dev/null | cut -d= -f2-)
fi

if [ -n "$VIOS_BASE_URL" ]; then
  curl -sf "${VIOS_BASE_URL}/vst/api/v1/sensor/list" >/dev/null \
    && echo "VIOS up at $VIOS_BASE_URL" \
    || { echo "VIOS_BASE_URL=$VIOS_BASE_URL is set but not responding"; VIOS_BASE_URL=""; }
fi

If VIOS is not reachable, ask the user to deploy VIOS and provide the base URL. Do not start RTSP capture until ${VIOS_BASE_URL}/vst/api/v1/sensor/list returns 200.

If VIOS is reachable but the AMC microservice is missing VIOS_BASE_URL, do not edit checked-in compose files. Export the variable and relaunch the microservice with a temporary compose override:

cd "$REPO_ROOT/compose"
export VIOS_BASE_URL="http://<VIOS_HOST>:30888"
OVERRIDE_FILE="${TMPDIR:-/tmp}/amc-vios.override.yml"
cat > "$OVERRIDE_FILE" <<'YAML'
services:
  auto-magic-calib-ms:
    environment:
      - VIOS_BASE_URL=${VIOS_BASE_URL}
YAML

docker compose -f compose.yml -f "$OVERRIDE_FILE" up -d auto-magic-calib-ms
docker exec auto-magic-calib-ms-1 printenv VIOS_BASE_URL

A host-shell export alone is not enough after the container is already running; the microservice process must be restarted with VIOS_BASE_URL in its environment.

Step 1 - Create Project

POST /v1/create_project with form field project_name. Save the returned project_id.

Step 2 - Start RTSP Capture

POST /v1/rtsp/capture/<project_id>
Content-Type: application/json

{
  "streams": [
    {"rtsp_url": "rtsp://...", "camera_name": "cam_00", "sensor_id": null},
    {"rtsp_url": "rtsp://...", "camera_name": "cam_01", "sensor_id": null}
  ],
  "duration_seconds": 180,
  "vios_token": null,
  "ssl_verify": false
}

The response can nest session fields under session:

{"code": 0, "message": "...", "session": {"session_id": "...", "status": "STARTING"}}

Save session.session_id.

Step 3 - Poll Capture, Then Ingest

Poll every 10 seconds:

GET /v1/rtsp/capture/<project_id>/<session_id>

Session lifecycle:

STARTING -> RECORDING -> COMPLETED -> INGESTING -> INGESTED
                       -> ERROR
RECORDING -> CANCELLED

When capture reaches COMPLETED, ingest the recorded clips into the AMC project:

POST /v1/rtsp/capture/<project_id>/<session_id>/ingest

After ingest succeeds, the project has video files attached and the rest of the workflow matches the MP4 upload path.

Need to stop early: POST /v1/rtsp/capture/<project_id>/<session_id>/stop. A partial clip can still be ingested if VIOS produced one.

Other session endpoints:

GET /v1/rtsp/sessions/<project_id> - list sessions for a project.
DELETE /v1/rtsp/session/<project_id>/<session_id> - delete a session record.

Step 4 - Upload Settings, Alignment, Layout, and Optional Files

Resolve local files using the anchor-file pattern above. Upload resolved files:

File	Endpoint	Notes
Calibration settings	`POST /v1/config/<project_id>`	JSON body posted as-is; replaces UI Step 3
Alignment JSON	`POST /v1/upload_alignment/<project_id>`	Multipart `alignment_file`
Layout PNG	`POST /v1/upload_layout/<project_id>`	Multipart `layout_file`
Ground truth zip	`POST /v1/upload_gt_file/<project_id>`	Optional
Focal lengths	`POST /v1/upload_focal_length/<project_id>`	Optional repeated `focal_length` values

Use only files from explicit user-provided paths or a user-provided calibration asset directory. Do not extract or scan sample data to find fallback settings, alignment, layout, or GT.

If settings are missing, direct the user to UI Step 3: Parameters at http://<HOST_IP>:<AUTO_MAGIC_CALIB_UI_PORT>, then ask which detector to use (resnet or transformer) before calibration. If alignment or layout is missing, direct the user to UI Step 4: Alignment for this project. For RTSP projects, videos are already ingested; do not re-upload videos in the UI fallback.

Before continuing after UI Step 4, verify:

PROJECT_ID=<project_id>
: "${REPO_ROOT:?set REPO_ROOT to the auto-magic-calib checkout. Run amc-setup-calibration-stack Step 0b first.}"
grep -q "AutoMagicCalib" "$REPO_ROOT/README.md" 2>/dev/null && grep -q "auto-magic-calib-ms" "$REPO_ROOT/compose/ms/compose.yml" 2>/dev/null || { echo "ERROR: REPO_ROOT is not an auto-magic-calib checkout: $REPO_ROOT" >&2; exit 1; }
PROJECT_DIR_REL=$(grep ^PROJECT_DIR "$REPO_ROOT/compose/.env" 2>/dev/null | cut -d= -f2 | tr -d '[:space:]')
HOST_PROJECTS=$(cd "$REPO_ROOT/compose" && realpath "${PROJECT_DIR_REL:-../../projects}")
ls "$HOST_PROJECTS/project_${PROJECT_ID}/manual_adjustment/"
# Expected: alignment_data.json, layout.png

Step 5 - Verify, Calibrate, Poll, and Fetch Results

Verify:

POST /v1/verify_project/<project_id>

The project must return project_state == "READY".

Confirm the plan before calibrating. Summarize:

Stream count and recording duration.
Detector: resnet or transformer.
Settings source: explicit uploaded settings file, user-provided asset directory, or UI Step 3.
Alignment/layout source: explicit uploaded files, user-provided asset directory, or UI manual adjustment.
Optional GT and focal-length overrides.

Start calibration:

POST /v1/calibrate/<project_id>
Content-Type: application/json

{"detector_type": "<resnet-or-transformer>"}

Poll:

GET /v1/get_project_info/<project_id>

Stop on COMPLETED or ERROR. On error, fetch GET /v1/amc/calibrate/<project_id>/log.

Fetch results:

GET /v1/result/<project_id>/evaluation_statistics

Only expect evaluation statistics when GT was uploaded.

Step 6 - Optional VGGT Refinement

After AMC calibration completes, read project_info.vggt_state from GET /v1/get_project_info/<project_id>.

If vggt_state == "READY", ask whether to run VGGT refinement.
If confirmed, call POST /v1/vggt/calibrate/<project_id>, poll vggt_state, then fetch GET /v1/vggt_results/<project_id>/evaluation_statistics.
If VGGT is not ready, skip it and explain that AMC calibration is complete.

Complete Python Script

Use the bundled script from the amc-run-rtsp-calibration skill package, not from the auto-magic-calib repo root. If the user points the agent at this skill folder directly instead of installing it, set AMC_RTSP_SKILL_DIR to the directory containing this SKILL.md, or run the command from that directory.

Common environment variables:

export BASE_URL=http://<HOST_IP>:8000
export PROJECT_NAME=rtsp_calibration_run
export RTSP_URLS='rtsp://user:pass@cam0/stream,rtsp://user:pass@cam1/stream'
export CAMERA_NAMES='cam_00,cam_01'
export DURATION_SECONDS=180
export VIOS_BASE_URL=http://<VIOS_HOST>:30888
export CALIB_ASSET_DIR=/path/to/rtsp-calibration-assets
# Or provide explicit CONFIG_FILE, ALIGNMENT_JSON, LAYOUT_PNG, and optional GT_ZIP.
export DETECTOR_TYPE=transformer  # Required when settings do not set detector/detector_type.
export AMC_UI_URL=http://<HOST_IP>:5000
export RUN_VGGT=false

# Optional but recommended: REPO_ROOT points to the auto-magic-calib checkout.
# PROJECTS_DIR can be set explicitly when project outputs live elsewhere.
if [ -z "${DEEPSTREAM_REPO_ROOT:-}" ] && [ -n "${REPO_ROOT:-}" ] && [ -d "$REPO_ROOT/../../skills/amc-run-rtsp-calibration" ]; then
  DEEPSTREAM_REPO_ROOT="$(cd "$REPO_ROOT/../.." && pwd)"
fi

SCRIPT_PATH=""
for candidate in \
  "${AMC_RTSP_SKILL_DIR:+$AMC_RTSP_SKILL_DIR/scripts/run_rtsp_calibration.py}" \
  "$PWD/scripts/run_rtsp_calibration.py" \
  "${DEEPSTREAM_REPO_ROOT:+$DEEPSTREAM_REPO_ROOT/skills/amc-run-rtsp-calibration/scripts/run_rtsp_calibration.py}" \
  "$PWD/skills/amc-run-rtsp-calibration/scripts/run_rtsp_calibration.py" \
  "$HOME/.claude/skills/amc-run-rtsp-calibration/scripts/run_rtsp_calibration.py" \
  "$HOME/.codex/skills/amc-run-rtsp-calibration/scripts/run_rtsp_calibration.py" \
  "$HOME/.cursor/skills/amc-run-rtsp-calibration/scripts/run_rtsp_calibration.py"; do
  if [ -f "$candidate" ]; then
    SCRIPT_PATH="$candidate"
    break
  fi
done

[ -n "$SCRIPT_PATH" ] || {
  echo "ERROR: could not find amc-run-rtsp-calibration/scripts/run_rtsp_calibration.py" >&2
  echo "Set AMC_RTSP_SKILL_DIR to the amc-run-rtsp-calibration skill directory, or run this block from that directory." >&2
  exit 1
}

python3 "$SCRIPT_PATH"

Alternative stream input:

export STREAMS_JSON='[
  {"rtsp_url":"rtsp://cam0/stream","camera_name":"cam_00","sensor_id":null},
  {"rtsp_url":"rtsp://cam1/stream","camera_name":"cam_01","sensor_id":null}
]'

Optional env vars are CALIB_ASSET_DIR, CONFIG_FILE, ALIGNMENT_JSON, LAYOUT_PNG, GT_ZIP, FOCAL_LENGTHS, DETECTOR_TYPE, AMC_UI_URL, VIOS_TOKEN, SSL_VERIFY, RUN_VGGT, REPO_ROOT, and PROJECTS_DIR.

Success Criteria

VIOS health probe returns 200.
Capture session reaches COMPLETED.
Ingest returns success and project info shows the expected video files.
verify_project returns READY.
AMC calibration reaches project_state == "COMPLETED".
If GT was uploaded, evaluation statistics are returned.
No RTSP credentials, bearer tokens, NGC keys, or HuggingFace tokens are printed or persisted by the agent.

Key Output Files

Results persist on the AMC server under:

projects/project_<project_id>/
|-- manual_adjustment/
|   |-- alignment_data.json
|   `-- layout.png
|-- output/
|   |-- single_view_results/cam_XX/
|   |   |-- camInfo_hyper_XX.yaml
|   |   `-- trajDump_Stream_0_3d.txt
|   `-- multi_view_results/BA_output/results_ba/
|       |-- initial/camInfo_XX.yaml
|       `-- refined/camInfo_XX.yaml
`-- calibration.log

Troubleshooting

Issue	Fix
VIOS `/vst/api/v1/sensor/list` returns connection refused	VIOS is not running or not reachable from this host. Ask the user to deploy VIOS or provide the reachable base URL.
Capture endpoint returns 503 or "VIOS not configured"	Export `VIOS_BASE_URL`, relaunch the microservice with the temporary compose override from Step 0, then retry capture.
Session stuck in `STARTING`	VIOS accepted the request but sensors may not be online. Check `${VIOS_BASE_URL}/vst/api/v1/sensor/list` and wait 20-30 seconds after sensor restarts.
Session stuck in `RECORDING` past `duration_seconds`	Call `POST /v1/rtsp/capture/<project_id>/<session_id>/stop`, then ingest the partial clip if available.
Ingest fails with "No clip available"	The recording window may not overlap the VIOS timeline. Wait for sensors to become online, then start a new capture.
400 "empty streams"	Pass at least one stream object with `rtsp_url` and `camera_name`.
400 "duration too short"	Use `duration_seconds >= 60`.
404 on `/v1/rtsp/capture/<project_id>`	Create the project first with `/v1/create_project`.
`verify_project` is not `READY` after ingest	Check project info and confirm expected videos, alignment, and layout are attached.
Calibration reaches `ERROR`	Fetch `GET /v1/amc/calibrate/<project_id>/log`; common causes are insufficient tracklets, static scenes, or incorrect alignment.

Related Skills

skills/amc-setup-calibration-stack/SKILL.md - start AMC microservice and UI first.
skills/amc-run-video-calibration/SKILL.md - calibrate from local pre-recorded MP4 files.
skills/amc-run-sample-calibration/SKILL.md - verify the stack with the bundled sample dataset.

When it activates

Automatically, when a task matches. Use when the user provides RTSP URLs or asks to calibrate live cameras; VIOS records clips, AMC ingests them, then runs calibration.

About this skill

Automatic

AI Engineer pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/17/2026

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

nvidia / skills

skills/amc-run-rtsp-calibration/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/17/2026Version v1Bundled 1 file · scripts

Illustrative Want to see what nvidia/amc-run-rtsp-calibration 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

Skill: Calibrate from RTSP Streams

When to Use This Skill

Prerequisites

Data Privacy

What to Ask the User

Required

Auto-Detected or Asked

Optional

Instructions

Step 0 - Verify AMC and VIOS

Step 1 - Create Project

Step 2 - Start RTSP Capture

Step 3 - Poll Capture, Then Ingest

Step 4 - Upload Settings, Alignment, Layout, and Optional Files

Step 5 - Verify, Calibrate, Poll, and Fetch Results

Step 6 - Optional VGGT Refinement

Complete Python Script

Success Criteria

Key Output Files

Troubleshooting

Related Skills

When it activates

SkillSafety

Source

Related skills

More in the registry