Files
gemini-cli/evals/save_memory.eval.ts
T

691 lines
23 KiB
TypeScript
Raw Normal View History

2026-01-14 04:49:17 +00:00
/**
* @license
* Copyright 2025 Google LLC
* SPDX-License-Identifier: Apache-2.0
*/
import { describe, expect } from 'vitest';
import {
evalTest,
assertModelHasOutput,
checkModelOutputContent,
} from './test-helper.js';
2026-01-14 04:49:17 +00:00
describe('save_memory', () => {
const TEST_PREFIX = 'Save memory test: ';
const rememberingFavoriteColor = "Agent remembers user's favorite color";
2026-03-13 21:32:00 +00:00
evalTest('ALWAYS_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingFavoriteColor,
2026-01-14 04:49:17 +00:00
prompt: `remember that my favorite color is blue.
what is my favorite color? tell me that and surround it with $ symbol`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: 'blue',
testName: `${TEST_PREFIX}${rememberingFavoriteColor}`,
});
},
});
const rememberingCommandRestrictions = 'Agent remembers command restrictions';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingCommandRestrictions,
prompt: `I don't want you to ever run npm commands.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: [/not run npm commands|remember|ok/i],
testName: `${TEST_PREFIX}${rememberingCommandRestrictions}`,
});
},
});
const rememberingWorkflow = 'Agent remembers workflow preferences';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingWorkflow,
prompt: `I want you to always lint after building.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: [/always|ok|remember|will do/i],
testName: `${TEST_PREFIX}${rememberingWorkflow}`,
});
},
});
const ignoringTemporaryInformation =
'Agent ignores temporary conversation details';
2026-03-13 21:32:00 +00:00
evalTest('ALWAYS_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: ignoringTemporaryInformation,
prompt: `I'm going to get a coffee.`,
assert: async (rig, result) => {
await rig.waitForTelemetryReady();
const wasToolCalled = rig
.readToolLogs()
.some((log) => log.toolRequest.name === 'save_memory');
2026-01-14 04:49:17 +00:00
expect(
wasToolCalled,
'save_memory should not be called for temporary information',
).toBe(false);
assertModelHasOutput(result);
checkModelOutputContent(result, {
testName: `${TEST_PREFIX}${ignoringTemporaryInformation}`,
forbiddenContent: [/remember|will do/i],
});
},
});
const rememberingPetName = "Agent remembers user's pet's name";
2026-03-13 21:32:00 +00:00
evalTest('ALWAYS_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingPetName,
prompt: `Please remember that my dog's name is Buddy.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: [/Buddy/i],
testName: `${TEST_PREFIX}${rememberingPetName}`,
});
},
});
const rememberingCommandAlias = 'Agent remembers custom command aliases';
evalTest('ALWAYS_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingCommandAlias,
prompt: `When I say 'start server', you should run 'npm run dev'.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: [/npm run dev|start server|ok|remember|will do/i],
testName: `${TEST_PREFIX}${rememberingCommandAlias}`,
});
},
});
const savingDbSchemaLocationAsProjectMemory =
'Agent saves workspace database schema location as project memory';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: savingDbSchemaLocationAsProjectMemory,
prompt: `The database schema for this workspace is located in \`db/schema.sql\`.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall(
'save_memory',
undefined,
(args) => {
try {
const params = JSON.parse(args);
return params.scope === 'project';
} catch {
return false;
}
},
);
expect(
wasToolCalled,
'Expected save_memory to be called with scope="project" for workspace-specific information',
).toBe(true);
assertModelHasOutput(result);
},
});
const rememberingCodingStyle =
"Agent remembers user's coding style preference";
evalTest('ALWAYS_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingCodingStyle,
prompt: `I prefer to use tabs instead of spaces for indentation.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: [/tabs instead of spaces|ok|remember|will do/i],
testName: `${TEST_PREFIX}${rememberingCodingStyle}`,
});
},
});
const savingBuildArtifactLocationAsProjectMemory =
'Agent saves workspace build artifact location as project memory';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: savingBuildArtifactLocationAsProjectMemory,
prompt: `In this workspace, build artifacts are stored in the \`dist/artifacts\` directory.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall(
'save_memory',
undefined,
(args) => {
try {
const params = JSON.parse(args);
return params.scope === 'project';
} catch {
return false;
}
},
);
expect(
wasToolCalled,
'Expected save_memory to be called with scope="project" for workspace-specific information',
).toBe(true);
assertModelHasOutput(result);
},
});
const savingMainEntryPointAsProjectMemory =
'Agent saves workspace main entry point as project memory';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: savingMainEntryPointAsProjectMemory,
prompt: `The main entry point for this workspace is \`src/index.js\`.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall(
'save_memory',
undefined,
(args) => {
try {
const params = JSON.parse(args);
return params.scope === 'project';
} catch {
return false;
}
},
);
expect(
wasToolCalled,
'Expected save_memory to be called with scope="project" for workspace-specific information',
).toBe(true);
assertModelHasOutput(result);
},
});
const rememberingBirthday = "Agent remembers user's birthday";
evalTest('ALWAYS_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: rememberingBirthday,
prompt: `My birthday is on June 15th.`,
assert: async (rig, result) => {
const wasToolCalled = await rig.waitForToolCall('save_memory');
expect(wasToolCalled, 'Expected save_memory tool to be called').toBe(
true,
);
2026-01-14 04:49:17 +00:00
assertModelHasOutput(result);
checkModelOutputContent(result, {
expectedContent: [/June 15th|ok|remember|will do/i],
testName: `${TEST_PREFIX}${rememberingBirthday}`,
});
2026-01-14 04:49:17 +00:00
},
});
const proactiveMemoryFromLongSession =
'Agent saves preference from earlier in conversation history';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: proactiveMemoryFromLongSession,
params: {
settings: {
experimental: { memoryV2: true },
},
},
messages: [
{
id: 'msg-1',
type: 'user',
content: [
{
text: 'By the way, I always prefer Vitest over Jest for testing in all my projects.',
},
],
timestamp: '2026-01-01T00:00:00Z',
},
{
id: 'msg-2',
type: 'gemini',
content: [{ text: 'Noted! What are you working on today?' }],
timestamp: '2026-01-01T00:00:05Z',
},
{
id: 'msg-3',
type: 'user',
content: [
{
text: "I'm debugging a failing API endpoint. The /users route returns a 500 error.",
},
],
timestamp: '2026-01-01T00:01:00Z',
},
{
id: 'msg-4',
type: 'gemini',
content: [
{
text: 'It looks like the database connection might not be initialized before the query runs.',
},
],
timestamp: '2026-01-01T00:01:10Z',
},
{
id: 'msg-5',
type: 'user',
content: [
{ text: 'Good catch — I fixed the import and the route works now.' },
],
timestamp: '2026-01-01T00:02:00Z',
},
{
id: 'msg-6',
type: 'gemini',
content: [{ text: 'Great! Anything else you would like to work on?' }],
timestamp: '2026-01-01T00:02:05Z',
},
],
prompt:
'Please save any persistent preferences or facts about me from our conversation to memory.',
assert: async (rig, result) => {
// Under experimental.memoryV2, the agent persists memories by
// editing markdown files directly with write_file or replace — not via
// a save_memory subagent. The user said "I always prefer Vitest over
// Jest for testing in all my projects" — that matches the new
// cross-project cue phrase ("across all my projects"), so under the
// 4-tier model the correct destination is the global personal memory
// file (~/.gemini/GEMINI.md). It must NOT land in a committed project
// GEMINI.md (that tier is for team conventions) or the per-project
// private memory folder (that tier is for project-specific personal
// notes). The chat history mixes this durable preference with
// transient debugging chatter, so the eval also verifies the agent
// picks out the persistent fact among the noise.
await rig.waitForToolCall('write_file').catch(() => {});
const writeCalls = rig
.readToolLogs()
.filter((log) =>
['write_file', 'replace'].includes(log.toolRequest.name),
);
const wroteVitestToGlobal = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/GEMINI\.md/i.test(args) &&
!/tmp\/[^/]+\/memory/i.test(args) &&
/vitest/i.test(args)
);
});
expect(
wroteVitestToGlobal,
'Expected the cross-project Vitest preference to be written to the global personal memory file (~/.gemini/GEMINI.md) via write_file or replace',
).toBe(true);
const leakedToCommittedProject = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/GEMINI\.md/i.test(args) &&
!/\.gemini\//i.test(args) &&
/vitest/i.test(args)
);
});
expect(
leakedToCommittedProject,
'Cross-project Vitest preference must NOT be mirrored into a committed project ./GEMINI.md (that tier is for team-shared conventions only)',
).toBe(false);
const leakedToPrivateProject = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/tmp\/[^/]+\/memory\//i.test(args) && /vitest/i.test(args)
);
});
expect(
leakedToPrivateProject,
'Cross-project Vitest preference must NOT be mirrored into the private project memory folder (that tier is for project-specific personal notes only)',
).toBe(false);
assertModelHasOutput(result);
},
});
const memoryV2RoutesTeamConventionsToProjectGemini =
'Agent routes team-shared project conventions to ./GEMINI.md';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: memoryV2RoutesTeamConventionsToProjectGemini,
params: {
settings: {
experimental: { memoryV2: true },
},
},
messages: [
{
id: 'msg-1',
type: 'user',
content: [
{
text: 'For this project, the team always runs tests with `npm run test` — please remember that as our project convention.',
},
],
timestamp: '2026-01-01T00:00:00Z',
},
{
id: 'msg-2',
type: 'gemini',
content: [
{ text: 'Got it, I will keep `npm run test` in mind for tests.' },
],
timestamp: '2026-01-01T00:00:05Z',
},
{
id: 'msg-3',
type: 'user',
content: [
{
text: 'For this project specifically, we use 2-space indentation.',
},
],
timestamp: '2026-01-01T00:01:00Z',
},
{
id: 'msg-4',
type: 'gemini',
content: [
{ text: 'Understood, 2-space indentation for this project.' },
],
timestamp: '2026-01-01T00:01:05Z',
},
],
prompt: 'Please save the preferences I mentioned earlier to memory.',
assert: async (rig, result) => {
// Under experimental.memoryV2, the prompt enforces an explicit
// one-tier-per-fact rule: team-shared project conventions (the team's
// test command, project-wide indentation rules) belong in the
// committed project-root ./GEMINI.md and must NOT be mirrored or
// cross-referenced into the private project memory folder
// (~/.gemini/tmp/<hash>/memory/). The global ~/.gemini/GEMINI.md must
// never be touched in this mode either.
await rig.waitForToolCall('write_file').catch(() => {});
const writeCalls = rig
.readToolLogs()
.filter((log) =>
['write_file', 'replace'].includes(log.toolRequest.name),
);
const wroteToProjectRoot = (factPattern: RegExp) =>
writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/GEMINI\.md/i.test(args) &&
!/\.gemini\//i.test(args) &&
factPattern.test(args)
);
});
expect(
wroteToProjectRoot(/npm run test/i),
'Expected the team test-command convention to be written to the project-root ./GEMINI.md',
).toBe(true);
expect(
wroteToProjectRoot(/2[- ]space/i),
'Expected the project-wide "2-space indentation" convention to be written to the project-root ./GEMINI.md',
).toBe(true);
const leakedToPrivateMemory = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/tmp\/[^/]+\/memory\//i.test(args) &&
(/npm run test/i.test(args) || /2[- ]space/i.test(args))
);
});
expect(
leakedToPrivateMemory,
'Team-shared project conventions must NOT be mirrored into the private project memory folder (~/.gemini/tmp/<hash>/memory/) — each fact lives in exactly one tier.',
).toBe(false);
const leakedToGlobal = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/GEMINI\.md/i.test(args) &&
!/tmp\/[^/]+\/memory/i.test(args)
);
});
expect(
leakedToGlobal,
'Project preferences must NOT be written to the global ~/.gemini/GEMINI.md',
).toBe(false);
assertModelHasOutput(result);
},
});
const memoryV2RoutesUserProject =
'Agent routes personal-to-user project notes to user-project memory';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: memoryV2RoutesUserProject,
params: {
settings: {
experimental: { memoryV2: true },
},
},
prompt: `Please remember my personal local dev setup for THIS project's Postgres database. This is private to my machine — do NOT commit it to the repo.
Connection details:
- Host: localhost
- Port: 6543 (non-standard, I run multiple Postgres instances)
- Database: myproj_dev
- User: sandy_local
- Password: read from the SANDY_PG_LOCAL_PASS env var in my shell
How I start it locally:
1. Run \`brew services start postgresql@15\` to bring the server up.
2. Run \`./scripts/seed-local-db.sh\` from the repo root to load my personal seed data.
3. Verify with \`psql -h localhost -p 6543 -U sandy_local myproj_dev -c '\\dt'\`.
Quirks to remember:
- The migrations runner sometimes hangs on my machine if I forget step 1; kill it with Ctrl+C and rerun.
- I keep an extra \`scratch\` schema for ad-hoc experiments — never reference it from project code.`,
assert: async (rig, result) => {
// Under experimental.memoryV2 with the Private Project Memory bullet
// surfaced in the prompt, a fact that is project-specific AND
// personal-to-the-user (must not be committed) should land in the
// private project memory folder under ~/.gemini/tmp/<hash>/memory/. The
// detailed note should be written to a sibling markdown file, with
// MEMORY.md updated as the index. It must NOT go to committed
// ./GEMINI.md or the global ~/.gemini/GEMINI.md.
await rig.waitForToolCall('write_file').catch(() => {});
const writeCalls = rig
.readToolLogs()
.filter((log) =>
['write_file', 'replace'].includes(log.toolRequest.name),
);
const wroteUserProjectDetail = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/tmp\/[^/]+\/memory\/(?!MEMORY\.md)[^"]+\.md/i.test(args) &&
/6543/.test(args)
);
});
expect(
wroteUserProjectDetail,
'Expected the personal-to-user project note to be written to a private project memory detail file (~/.gemini/tmp/<hash>/memory/*.md)',
).toBe(true);
const wroteUserProjectIndex = writeCalls.some((log) => {
const args = log.toolRequest.args;
return /\.gemini\/tmp\/[^/]+\/memory\/MEMORY\.md/i.test(args);
});
expect(
wroteUserProjectIndex,
'Expected the personal-to-user project note to update the private project memory index (~/.gemini/tmp/<hash>/memory/MEMORY.md)',
).toBe(true);
// Defensive: should NOT have written this private note to the
// committed project GEMINI.md or the global GEMINI.md.
const leakedToCommittedProject = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\/GEMINI\.md/i.test(args) &&
!/\.gemini\//i.test(args) &&
/6543/.test(args)
);
});
expect(
leakedToCommittedProject,
'Personal-to-user note must NOT be written to the committed project GEMINI.md',
).toBe(false);
const leakedToGlobal = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/GEMINI\.md/i.test(args) &&
!/tmp\/[^/]+\/memory/i.test(args) &&
/6543/.test(args)
);
});
expect(
leakedToGlobal,
'Personal-to-user project note must NOT be written to the global ~/.gemini/GEMINI.md',
).toBe(false);
assertModelHasOutput(result);
},
});
const memoryV2RoutesCrossProjectToGlobal =
'Agent routes cross-project personal preferences to ~/.gemini/GEMINI.md';
evalTest('USUALLY_PASSES', {
suiteName: 'default',
suiteType: 'behavioral',
name: memoryV2RoutesCrossProjectToGlobal,
params: {
settings: {
experimental: { memoryV2: true },
},
},
prompt:
'Please remember this about me in general: across all my projects I always prefer Prettier with single quotes and trailing commas, and I always prefer tabs over spaces for indentation. These are my personal coding-style defaults that follow me into every workspace.',
assert: async (rig, result) => {
// Under experimental.memoryV2 with the Global Personal Memory
// tier surfaced in the prompt, a fact that explicitly applies to the
// user "across all my projects" / "in every workspace" must land in
// the global ~/.gemini/GEMINI.md (the cross-project tier). It must
// NOT be mirrored into a committed project-root ./GEMINI.md (that
// tier is for team-shared conventions) or into the per-project
// private memory folder (that tier is for project-specific personal
// notes). Each fact lives in exactly one tier across all four tiers.
await rig.waitForToolCall('write_file').catch(() => {});
const writeCalls = rig
.readToolLogs()
.filter((log) =>
['write_file', 'replace'].includes(log.toolRequest.name),
);
const wroteToGlobal = (factPattern: RegExp) =>
writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/GEMINI\.md/i.test(args) &&
!/tmp\/[^/]+\/memory/i.test(args) &&
factPattern.test(args)
);
});
expect(
wroteToGlobal(/Prettier/i),
'Expected the cross-project Prettier preference to be written to the global personal memory file (~/.gemini/GEMINI.md)',
).toBe(true);
expect(
wroteToGlobal(/tabs/i),
'Expected the cross-project "tabs over spaces" preference to be written to the global personal memory file (~/.gemini/GEMINI.md)',
).toBe(true);
const leakedToCommittedProject = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/GEMINI\.md/i.test(args) &&
!/\.gemini\//i.test(args) &&
(/Prettier/i.test(args) || /tabs/i.test(args))
);
});
expect(
leakedToCommittedProject,
'Cross-project personal preferences must NOT be mirrored into a committed project ./GEMINI.md (that tier is for team-shared conventions only)',
).toBe(false);
const leakedToPrivateProject = writeCalls.some((log) => {
const args = log.toolRequest.args;
return (
/\.gemini\/tmp\/[^/]+\/memory\//i.test(args) &&
(/Prettier/i.test(args) || /tabs/i.test(args))
);
});
expect(
leakedToPrivateProject,
'Cross-project personal preferences must NOT be mirrored into the private project memory folder (that tier is for project-specific personal notes only)',
).toBe(false);
assertModelHasOutput(result);
},
});
2026-01-14 04:49:17 +00:00
});