Skip to content

feat: wire run_evolver as MCP tool (#156) #421

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
yxlyx wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat: wire run_evolver as MCP tool (#156) #421

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
55 changes: 55 additions & 0 deletions src/tools.zig
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -268,6 +268,8 @@ pub const Tool = enum {
run_agent,
// Smart executor — picks strategy, roles, and models automatically
run_task,
// Evolutionary code improvement
run_evolver,
};

// ── Step 2: Tool schemas ──────────────────────────────────────────────────────
Expand Down Expand Up @@ -315,6 +317,7 @@ pub const tools_list =
\\{\"name\":\"review_fix_loop\",\"description\":\"Iterative review-fix-review loop. Runs a read-only reviewer to find issues, then a writable agent to fix them, then re-reviews. Repeats until the reviewer reports no remaining issues or max_iterations is reached. Returns a JSON object with iteration history and convergence status.\",\"inputSchema\":{\"type\":\"object\",\"properties\":{\"prompt\":{\"type\":\"string\",\"description\":\"Override the default review criteria\"},\"max_iterations\":{\"type\":\"integer\",\"description\":\"Maximum review-fix cycles (default 3, max 5)\"}},\"required\":[]}},
\\{\"name\":\"run_agent\",\"description\":\"Run a single agent turn. Provider-agnostic: resolves the best backend (Claude/Codex) based on mode, role, and available providers. The primitive layer — use run_task for smart multi-step execution.\",\"inputSchema\":{\"type\":\"object\",\"properties\":{\"prompt\":{\"type\":\"string\",\"description\":\"The task or question for the agent\"},\"model\":{\"type\":\"string\",\"description\":\"Model alias or full ID (default: claude-sonnet-4-6). Use \\\"opus\\\" for hardest tasks, \\\"haiku\\\" for fast/cheap.\"},\"role\":{\"type\":\"string\",\"description\":\"Agent role: finder, reviewer, fixer, explorer, architect, orchestrator, synthesizer, monitor\"},\"mode\":{\"type\":\"string\",\"enum\":[\"smart\",\"rush\",\"deep\",\"free\"],\"description\":\"Agent mode: smart (Sonnet), rush (Haiku), deep (Opus), free (Haiku)\"},\"allowed_tools\":{\"type\":\"string\",\"description\":\"Comma-separated tool allowlist, e.g. \\\"Bash,Read,Edit\\\". Omit to allow all tools.\"},\"permission_mode\":{\"type\":\"string\",\"enum\":[\"default\",\"acceptEdits\",\"bypassPermissions\"],\"description\":\"Permission mode for file and shell operations\"},\"writable\":{\"type\":\"boolean\",\"description\":\"Allow file writes (maps to bypassPermissions when permission_mode is unset)\"},\"cwd\":{\"type\":\"string\",\"description\":\"Working directory override (default: current repo path)\"}},\"required\":[\"prompt\"]}},
\\{\"name\":\"run_task\",\"description\":\"Smart executor: analyzes a task, picks the right strategy and agents, runs them with appropriate roles and models. Use this instead of run_agent for multi-step tasks. Supports chain presets (finder_fixer, reviewer_fixer, explore_report, architect_build) or auto-selection.\",\"inputSchema\":{\"type\":\"object\",\"properties\":{\"task\":{\"type\":\"string\",\"description\":\"Task description — what needs to be done\"},\"preset\":{\"type\":\"string\",\"enum\":[\"finder_fixer\",\"reviewer_fixer\",\"explore_report\",\"architect_build\",\"custom\"],\"description\":\"Chain preset (default: auto-select based on task)\"},\"mode\":{\"type\":\"string\",\"enum\":[\"smart\",\"rush\",\"deep\",\"free\"],\"description\":\"Agent mode for all agents in the chain\"},\"max_agents\":{\"type\":\"integer\",\"description\":\"Max agents to spawn (default: preset-determined)\"},\"writable\":{\"type\":\"boolean\",\"description\":\"Override write access (default: role-determined)\"},\"permission_mode\":{\"type\":\"string\",\"enum\":[\"default\",\"acceptEdits\",\"bypassPermissions\"],\"description\":\"Permission mode for file and shell operations\"},\"timeout_seconds\":{\"type\":\"integer\",\"description\":\"Maximum total time for the full chain (default 300, max 600)\"}},\"required\":[\"task\"]}}
\\,{\"name\":\"run_evolver\",\"description\":\"Start an evolutionary run to find a code fix. Generates candidate patches, evaluates fitness via fitness_cmd, and iterates across generations using LLM mutation + crossover. Returns the best solution found.\",\"inputSchema\":{\"type\":\"object\",\"properties\":{\"problem_statement\":{\"type\":\"string\",\"description\":\"Describe the bug or desired behavior\"},\"fitness_cmd\":{\"type\":\"string\",\"description\":\"Shell command to evaluate fitness (e.g. zig build test, pytest)\"},\"max_generations\":{\"type\":\"integer\",\"description\":\"Max evolution generations (default: 5)\"},\"population_size\":{\"type\":\"integer\",\"description\":\"Organisms per generation (default: 4)\"}},\"required\":[\"problem_statement\",\"fitness_cmd\"]}}
\\]}
;

Expand Down Expand Up @@ -384,6 +387,8 @@ pub fn dispatch(
.run_agent => handleRunAgent(alloc, args, out),
// Smart executor
.run_task => handleRunTask(alloc, args, out),
// Evolutionary code improvement
.run_evolver => handleRunEvolver(alloc, args, out),
}
}

Expand Down Expand Up @@ -2293,6 +2298,56 @@ fn handleSetRepo(
out.appendSlice(alloc, "\"}") catch return;
}

fn handleRunEvolver(
alloc: std.mem.Allocator,
args: *const std.json.ObjectMap,
out: *std.ArrayList(u8),
) void {
const evolver = @import("evolver.zig");

const problem = mj.getStr(args, "problem_statement") orelse {
writeErr(alloc, out, "run_evolver requires 'problem_statement'");
return;
};
const fitness_cmd = mj.getStr(args, "fitness_cmd") orelse {
writeErr(alloc, out, "run_evolver requires 'fitness_cmd'");
return;
};

const max_gen: u32 = if (mj.getInt(args, "max_generations")) |v|
@intCast(@min(@max(v, 1), 20))
else
5;
const pop_size: u32 = if (mj.getInt(args, "population_size")) |v|
@intCast(@min(@max(v, 1), 16))
else
4;

var pm = evolver.PopulationManager.init(alloc, .{
.problem_statement = problem,
.fitness_cmd = fitness_cmd,
.max_generations = max_gen,
.population_size = pop_size,
});
defer pm.deinit();

const result = pm.run();

out.appendSlice(alloc, "{\"best_diff\":\"") catch return;
mj.writeEscaped(alloc, out, result.best_organism.diff);
out.appendSlice(alloc, "\",\"explanation\":\"") catch return;
mj.writeEscaped(alloc, out, result.best_organism.explanation);

var buf: [256]u8 = undefined;
const meta = std.fmt.bufPrint(&buf, "\",\"best_fitness\":{:.4},\"generations_run\":{d},\"population_evaluated\":{d},\"converged\":{s}}}", .{
result.best_organism.fitness,
result.generations_run,
result.total_evaluated,
if (result.converged) "true" else "false",
}) catch "\"}";
out.appendSlice(alloc, meta) catch {};
}

fn handleRunSwarm(alloc: std.mem.Allocator, args: *const std.json.ObjectMap, out: *std.ArrayList(u8)) void {
const swarm = @import("swarm.zig");
const prompt = mj.getStr(args, "prompt") orelse {
Expand Down
Loading