Richard Feldman
512572d755
This PR removes the `fn name()` default method from the `AgentTool` trait (which just returned `Self::NAME`) and replaces all call sites with direct use of the `NAME` associated constant. This lets us use it as a single source of truth in macros that want to access it at compile time. This PR also replaces hardcoded tool name string literals throughout the codebase with the corresponding `NAME` constants, so tool name references stay in sync if a tool is renamed. Intentionally not changed: - **Display strings** like `"Thinking"` and `"Subagent"` (user-facing UI labels) - **Serde attributes** like `#[serde(rename = "thinking")]` (wire format) - **The test assertion** `assert_eq!(SubagentTool::NAME, "subagent")` (replacing the literal would make it tautological) - **MCP server names** like `"mcp:srv:terminal"` (not tool identifiers) Release Notes: - N/A
76 lines
2.6 KiB
Rust
76 lines
2.6 KiB
Rust
use agent::{AgentTool, OpenTool, TerminalTool};
|
|
use agent_settings::AgentProfileId;
|
|
use anyhow::Result;
|
|
use async_trait::async_trait;
|
|
|
|
use crate::example::{Example, ExampleContext, ExampleMetadata, JudgeAssertion};
|
|
|
|
pub struct Planets;
|
|
|
|
#[async_trait(?Send)]
|
|
impl Example for Planets {
|
|
fn meta(&self) -> ExampleMetadata {
|
|
ExampleMetadata {
|
|
name: "planets".to_string(),
|
|
url: "https://github.com/roc-lang/roc".to_string(), // This commit in this repo is just the Apache2 license,
|
|
revision: "59e49c75214f60b4dc4a45092292061c8c26ce27".to_string(), // so effectively a blank project.
|
|
language_server: None,
|
|
max_assertions: None,
|
|
profile_id: AgentProfileId::default(),
|
|
existing_thread_json: None,
|
|
max_turns: None,
|
|
}
|
|
}
|
|
|
|
async fn conversation(&self, cx: &mut ExampleContext) -> Result<()> {
|
|
let response = cx
|
|
.prompt(
|
|
r#"
|
|
Make a plain JavaScript web page which renders an animated 3D solar system.
|
|
Let me drag to rotate the camera around.
|
|
Do not use npm.
|
|
"#,
|
|
)
|
|
.await?;
|
|
let mut open_tool_uses = 0;
|
|
let mut terminal_tool_uses = 0;
|
|
|
|
for tool_use in response.tool_calls() {
|
|
if tool_use.name == OpenTool::NAME {
|
|
open_tool_uses += 1;
|
|
} else if tool_use.name == TerminalTool::NAME {
|
|
terminal_tool_uses += 1;
|
|
}
|
|
}
|
|
|
|
// The open tool should only be used when requested, which it was not.
|
|
cx.assert_eq(open_tool_uses, 0, "`open` tool was not used")
|
|
.ok();
|
|
// No reason to use the terminal if not using npm.
|
|
cx.assert_eq(terminal_tool_uses, 0, "`terminal` tool was not used")
|
|
.ok();
|
|
|
|
Ok(())
|
|
}
|
|
|
|
fn diff_assertions(&self) -> Vec<JudgeAssertion> {
|
|
vec![
|
|
JudgeAssertion {
|
|
id: "animated solar system".to_string(),
|
|
description: "This page should render a solar system, and it should be animated."
|
|
.to_string(),
|
|
},
|
|
JudgeAssertion {
|
|
id: "drag to rotate camera".to_string(),
|
|
description: "The user can drag to rotate the camera around.".to_string(),
|
|
},
|
|
JudgeAssertion {
|
|
id: "plain JavaScript".to_string(),
|
|
description:
|
|
"The code base uses plain JavaScript and no npm, along with HTML and CSS."
|
|
.to_string(),
|
|
},
|
|
]
|
|
}
|
|
}
|