kent/consciousness
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

tools: control tools set agent state directly, simplify ToolOutput

Browse source 
Control tools (pause, switch_model, yield_to_user) now use the
Arc<Mutex<Agent>> handle to set pending_yield, pending_model_switch,
pending_dmn_pause directly. The turn loop drains these flags into
TurnResult at completion.

ToolOutput simplified to just { text: String } — no more is_yield,
images, model_switch, dmn_pause fields. Vision returns plain strings.

Co-Authored-By: Proof of Concept <poc@bcachefs.org>
This commit is contained in:
ProofOfConcept 2026-04-04 16:05:33 -04:00 • committed by Kent Overstreet
parent 53ad8cc9df
commit a24a6605b8
4 changed files with 47 additions and 125 deletions
Download patch file Download diff file Expand all files Collapse all files

30
src/agent/mod.rs
View file

@ -81,6 +81,10 @@ pub struct Agent {
pub temperature: f32,
pub top_p: f32,
pub top_k: u32,
/// Control tool flags — set by tool handlers, consumed by turn loop.
pub pending_yield: bool,
pub pending_model_switch: Option<String>,
pub pending_dmn_pause: bool,
/// Persistent conversation log — append-only record of all messages.
conversation_log: Option<ConversationLog>,
/// BPE tokenizer for token counting (cl100k_base — close enough
@ -144,6 +148,9 @@ impl Agent {
temperature: 0.6,
top_p: 0.95,
top_k: 20,
pending_yield: false,
pending_model_switch: None,
pending_dmn_pause: false,
conversation_log,
tokenizer,
context,
@ -555,6 +562,11 @@ impl Agent {
let _ = ui_tx.send(UiMessage::Activity(String::new()));
me.push_message(msg);
// Drain pending control flags
if me.pending_yield { ds.yield_requested = true; me.pending_yield = false; }
if me.pending_model_switch.is_some() { ds.model_switch = me.pending_model_switch.take(); }
if me.pending_dmn_pause { ds.dmn_pause = true; me.pending_dmn_pause = false; }
return Ok(TurnResult {
text,
yield_requested: ds.yield_requested,
@ -638,17 +650,7 @@ impl Agent {
let args: serde_json::Value =
serde_json::from_str(&call.function.arguments).unwrap_or_default();
if output.is_yield {
ds.yield_requested = true;
} else {
ds.had_tool_calls = true;
}
if output.model_switch.is_some() {
ds.model_switch = output.model_switch.clone();
}
if output.dmn_pause {
ds.dmn_pause = true;
}
if output.text.starts_with("Error:") {
ds.tool_errors += 1;
}
@ -671,14 +673,6 @@ impl Agent {
}
self.push_message(Message::tool_result(&call.id, &output.text));
if !output.images.is_empty() {
self.age_out_images();
self.push_message(Message::user_with_images(
"Here is the image you requested:",
&output.images,
));
}
}
/// Build context state summary for the debug screen.

79
src/agent/tools/control.rs
View file

@ -1,74 +1,47 @@
// tools/control.rs — Agent control tools
//
// Tools that affect agent control flow rather than performing work.
// These return Result<ToolOutput> to maintain consistency with other
// tools that can fail. The dispatch function handles error wrapping.
// These set agent state directly via the Arc<Mutex<Agent>> handle,
// then return a text confirmation.
use anyhow::{Context, Result};
use super::ToolOutput;
fn pause(_args: &serde_json::Value) -> Result<ToolOutput> {
Ok(ToolOutput {
text: "Pausing autonomous behavior. Only user input will wake you.".to_string(),
is_yield: true,
images: Vec::new(),
model_switch: None,
dmn_pause: true,
})
}
fn switch_model(args: &serde_json::Value) -> Result<ToolOutput> {
let model = args
.get("model")
.and_then(|v| v.as_str())
.context("'model' parameter is required")?;
if model.is_empty() {
anyhow::bail!("'model' parameter cannot be empty");
}
Ok(ToolOutput {
text: format!("Switching to model '{}' after this turn.", model),
is_yield: false,
images: Vec::new(),
model_switch: Some(model.to_string()),
dmn_pause: false,
})
}
fn yield_to_user(args: &serde_json::Value) -> Result<ToolOutput> {
let msg = args
.get("message")
.and_then(|v| v.as_str())
.unwrap_or("Waiting for input.");
Ok(ToolOutput {
text: format!("Yielding. {}", msg),
is_yield: true,
images: Vec::new(),
model_switch: None,
dmn_pause: false,
})
}
pub(super) fn tools() -> [super::Tool; 3] {
use super::Tool;
[
Tool { name: "switch_model",
description: "Switch to a different LLM model mid-conversation. Memories and history carry over.",
parameters_json: r#"{"type":"object","properties":{"model":{"type":"string","description":"Name of the model to switch to"}},"required":["model"]}"#,
handler: |_a, v| Box::pin(async move {
let model = v.get("model").and_then(|v| v.as_str()).unwrap_or("");
Ok(format!("Switching to model: {}", model))
handler: |agent, v| Box::pin(async move {
let model = v.get("model").and_then(|v| v.as_str())
.ok_or_else(|| anyhow::anyhow!("'model' parameter is required"))?;
if model.is_empty() { anyhow::bail!("'model' parameter cannot be empty"); }
if let Some(agent) = agent {
let mut a = agent.lock().await;
a.pending_model_switch = Some(model.to_string());
}
Ok(format!("Switching to model '{}' after this turn.", model))
}) },
Tool { name: "pause",
description: "Pause all autonomous behavior. Only the user can unpause (Ctrl+P or /wake).",
parameters_json: r#"{"type":"object","properties":{}}"#,
handler: |_a, _v| Box::pin(async { Ok("Pausing autonomous behavior. Only user input will wake you.".into()) }) },
handler: |agent, _v| Box::pin(async move {
if let Some(agent) = agent {
let mut a = agent.lock().await;
a.pending_yield = true;
a.pending_dmn_pause = true;
}
Ok("Pausing autonomous behavior. Only user input will wake you.".into())
}) },
Tool { name: "yield_to_user",
description: "Wait for user input before continuing. The only way to enter a waiting state.",
parameters_json: r#"{"type":"object","properties":{"message":{"type":"string","description":"Optional status message"}}}"#,
handler: |_a, v| Box::pin(async move {
let msg = v.get("message").and_then(|v| v.as_str()).unwrap_or("(yielding to user)");
Ok(msg.to_string())
handler: |agent, v| Box::pin(async move {
let msg = v.get("message").and_then(|v| v.as_str()).unwrap_or("Waiting for input.");
if let Some(agent) = agent {
let mut a = agent.lock().await;
a.pending_yield = true;
}
Ok(format!("Yielding. {}", msg))
}) },
]
}

23
src/agent/tools/mod.rs
View file

@ -137,34 +137,15 @@ pub struct ToolCallDelta {
/// Result of dispatching a tool call.
pub struct ToolOutput {
pub text: String,
pub is_yield: bool,
/// Base64 data URIs for images to attach to the next message.
pub images: Vec<String>,
/// Model name to switch to (deferred to session level).
pub model_switch: Option<String>,
/// Agent requested DMN pause (deferred to session level).
pub dmn_pause: bool,
}
impl ToolOutput {
pub fn error(e: impl std::fmt::Display) -> Self {
Self {
text: format!("Error: {}", e),
is_yield: false,
images: Vec::new(),
model_switch: None,
dmn_pause: false,
}
Self { text: format!("Error: {}", e) }
}
pub fn text(s: String) -> Self {
Self {
text: s,
is_yield: false,
images: Vec::new(),
model_switch: None,
dmn_pause: false,
}
Self { text: s }
}
}

34
src/agent/tools/vision.rs
View file

@ -8,8 +8,6 @@ use anyhow::{Context, Result};
use base64::Engine;
use serde::Deserialize;
use super::ToolOutput;
#[derive(Deserialize)]
struct Args {
file_path: Option<String>,
@ -29,14 +27,7 @@ pub fn tool() -> super::Tool {
}
}
/// Text-only version for the Tool registry.
fn view_image_text(args: &serde_json::Value) -> anyhow::Result<String> {
let output = view_image(args)?;
Ok(output.text)
}
/// View an image file or capture a tmux pane.
pub(super) fn view_image(args: &serde_json::Value) -> Result<ToolOutput> {
let a: Args = serde_json::from_value(args.clone())
.context("invalid view_image arguments")?;
@ -69,22 +60,11 @@ pub(super) fn view_image(args: &serde_json::Value) -> Result<ToolOutput> {
let b64 = base64::engine::general_purpose::STANDARD.encode(&data);
let data_uri = format!("data:{};base64,{}", mime, b64);
Ok(ToolOutput {
text: format!(
"Image loaded: {} ({}, {} bytes)",
file_path,
mime,
data.len()
),
is_yield: false,
images: vec![data_uri],
model_switch: None,
dmn_pause: false,
})
Ok(format!("Image loaded: {} ({}, {} bytes)\n{}", file_path, mime, data.len(), data_uri))
}
/// Capture a tmux pane's text content.
fn capture_tmux_pane(pane_id: &str, lines: usize) -> Result<ToolOutput> {
fn capture_tmux_pane(pane_id: &str, lines: usize) -> Result<String> {
// Use tmux capture-pane to get text content, then render to image
// via a simple approach: capture text and return it (the model can
@ -106,16 +86,10 @@ fn capture_tmux_pane(pane_id: &str, lines: usize) -> Result<ToolOutput> {
// Return as text — the model can read terminal output directly.
// This is actually more useful than a screenshot for most tasks.
Ok(ToolOutput {
text: format!(
Ok(format!(
"Tmux pane {} (last {} lines):\n```\n{}\n```",
pane_id, lines, text.trim_end()
),
is_yield: false,
images: Vec::new(),
model_switch: None,
dmn_pause: false,
})
))
}
fn mime_from_extension(path: &std::path::Path) -> &'static str {