Explicit Tool Calls¶

OneTool's >>> prefix gives you explicit control over tool invocation. You write the code — the agent doesn't have to guess which tool or which parameters.

Snippet Mode vs Code Mode¶

Snippet Mode¶

Jinja2 templates invoked with $name. Values are plain strings; Python syntax does not apply.

>>> $g q=latest AI tools
>>> $pkg_npm packages=react lodash
>>> $g q="AI news"

Quotes are optional and stripped (q=abc ≡ q="abc")
Param names support prefix abbreviation (q resolves to query if defined)
Per-template features (e.g. pipe batch) are not snippet language features

Code Mode¶

Python executed directly against the tool namespace.

>>> brave.search(q="AI news")
>>> x = foo(text="hello"); x

Python syntax applies: strings must be quoted
Short param names work: q resolves to query (pack proxy prefix resolution)
Keyword arguments only: fn(key="val") not fn("val")

Trigger Hierarchy¶

Prefix	Role
`>>>`	Recommended; Python REPL symbol
`__run`	Systematic short form (`__(tool)`)
`mcp__onetool__run`	Canonical MCP name
`__ot`, `__onetool`	Deprecated — kept for backward compat, use `>>>` instead

Note: mcp__ot__run is NOT a valid prefix.

Invocation Styles¶

Simple Call¶

Direct function call after the prefix:

>>> foo(text="hello")
>>> multiply(a=8472, b=9384)

Code Fence¶

Multi-line code in a fenced block:

>>>
```python
metals = ["Gold", "Silver"]
results = {}
for metal in metals:
    results[metal] = brave.search(query=f"{metal} price")
results
```

Direct MCP Call¶

For programmatic or explicit MCP invocation:

mcp__onetool__run foo(text="hello")

Complete Examples¶

Simple Hash Computation¶

>>> foo(text="hello world")

Multi-step Computation with Variables¶

>>>
```python
msg = "Hello World"
foo(text=msg)
```

Loop with Multiple Tool Calls¶

>>>
```python
primes = [is_prime(n=i) for i in range(11, 21)]
primes
```

Prompt Engineering for Tool Calls¶

Pre-Call Instructions¶

Add context before the tool call to guide the agent:

Process the following text:
>>> foo(text="hello world")

Post-Call Processing¶

Request specific formatting or analysis after the tool result:

>>> brave.search(query="latest AI news", count=5)

Summarise the top 3 results in bullet points.

Structured Output Requests¶

Combine tool execution with output formatting:

>>>
```python
results = {
    "hash": foo(text="hello"),
    "reversed": reverse(text="hello"),
    "length": count_chars(text="hello")
}
results
```

Return the results as a markdown table.