Files
flynn/src/tools/integration.test.ts
T
William Valentin 90ce622080 feat(policy): enforce truthfulness and autonomy guardrails
Add runtime truthfulness modes and autonomy-level tool gating with audit metadata for overrides/denials.

Wire policy through prompt assembly, tool execution context, and daemon/gateway agent paths; update tests and planning state for Phase 3 PR #2 completion.
2026-02-12 16:06:45 -08:00

153 lines
5.3 KiB
TypeScript

// src/tools/integration.test.ts
import { describe, it, expect, vi } from 'vitest';
import { NativeAgent } from '../backends/native/agent.js';
import { ToolRegistry } from './registry.js';
import { ToolExecutor } from './executor.js';
import { HookEngine } from '../hooks/engine.js';
import { shellExecTool } from './builtin/shell.js';
import { fileReadTool } from './builtin/file-read.js';
import { fileWriteTool } from './builtin/file-write.js';
import type { ModelClient, ChatResponse } from '../models/types.js';
import { mkdtempSync, rmSync } from 'fs';
import { join } from 'path';
import { tmpdir } from 'os';
describe('Tool integration (end-to-end)', () => {
it('agent uses shell tool and returns result', async () => {
let callCount = 0;
const mockClient: ModelClient = {
chat: vi.fn().mockImplementation(() => {
callCount++;
if (callCount === 1) {
return {
content: '',
stopReason: 'tool_use',
usage: { inputTokens: 10, outputTokens: 5 },
toolCalls: [{ id: 'c1', name: 'shell.exec', args: { command: 'echo integration_test' } }],
} satisfies ChatResponse;
}
return {
content: 'The command output was: integration_test',
stopReason: 'end_turn',
usage: { inputTokens: 20, outputTokens: 10 },
} satisfies ChatResponse;
}),
};
const registry = new ToolRegistry();
registry.register(shellExecTool);
const hooks = new HookEngine({ confirm: [], log: [], silent: [] });
const executor = new ToolExecutor(registry, hooks);
const agent = new NativeAgent({
modelClient: mockClient,
systemPrompt: 'You have tools.',
toolRegistry: registry,
toolExecutor: executor,
toolPolicyContext: { autonomyLevel: 'autonomous' },
});
const result = await agent.process('run echo integration_test');
expect(result).toContain('integration_test');
});
it('agent chains multiple tools across iterations', async () => {
const dir = mkdtempSync(join(tmpdir(), 'flynn-integ-'));
let callCount = 0;
const mockClient: ModelClient = {
chat: vi.fn().mockImplementation(() => {
callCount++;
if (callCount === 1) {
return {
content: '',
stopReason: 'tool_use',
usage: { inputTokens: 10, outputTokens: 5 },
toolCalls: [{ id: 'c1', name: 'file.write', args: { path: join(dir, 'test.txt'), content: 'hello' } }],
};
}
if (callCount === 2) {
return {
content: '',
stopReason: 'tool_use',
usage: { inputTokens: 15, outputTokens: 8 },
toolCalls: [{ id: 'c2', name: 'file.read', args: { path: join(dir, 'test.txt') } }],
};
}
return {
content: 'I wrote and read the file. It contains: hello',
stopReason: 'end_turn',
usage: { inputTokens: 20, outputTokens: 10 },
};
}),
};
const registry = new ToolRegistry();
registry.register(fileWriteTool);
registry.register(fileReadTool);
const hooks = new HookEngine({ confirm: [], log: [], silent: [] });
const executor = new ToolExecutor(registry, hooks);
const agent = new NativeAgent({
modelClient: mockClient,
systemPrompt: 'You have tools.',
toolRegistry: registry,
toolExecutor: executor,
toolPolicyContext: { autonomyLevel: 'autonomous' },
});
try {
const result = await agent.process('write hello to test.txt then read it');
expect(result).toContain('hello');
expect(mockClient.chat).toHaveBeenCalledTimes(3);
} finally {
rmSync(dir, { recursive: true });
}
});
it('verifies tool results are passed back to model correctly', async () => {
let callCount = 0;
const mockClient: ModelClient = {
chat: vi.fn().mockImplementation((request: { messages: unknown[] }) => {
callCount++;
if (callCount === 1) {
return {
content: '',
stopReason: 'tool_use',
usage: { inputTokens: 10, outputTokens: 5 },
toolCalls: [{ id: 'c1', name: 'shell.exec', args: { command: 'echo verify_pass' } }],
};
}
// Second call: verify the tool result was included in messages
const lastMsg = request.messages[request.messages.length - 1] as { content: unknown[] };
const resultBlock = lastMsg.content[0] as { type: string; tool_use_id: string; content: string };
expect(resultBlock.type).toBe('tool_result');
expect(resultBlock.tool_use_id).toBe('c1');
expect(resultBlock.content).toContain('verify_pass');
return {
content: 'Verified tool result',
stopReason: 'end_turn',
usage: { inputTokens: 20, outputTokens: 10 },
};
}),
};
const registry = new ToolRegistry();
registry.register(shellExecTool);
const hooks = new HookEngine({ confirm: [], log: [], silent: [] });
const executor = new ToolExecutor(registry, hooks);
const agent = new NativeAgent({
modelClient: mockClient,
systemPrompt: 'You have tools.',
toolRegistry: registry,
toolExecutor: executor,
toolPolicyContext: { autonomyLevel: 'autonomous' },
});
const result = await agent.process('verify tool results');
expect(result).toBe('Verified tool result');
});
});