-
Notifications
You must be signed in to change notification settings - Fork 46
Add prompt-caching examples for Effect AI #44
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.
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
Merged
Merged
Changes from all commits
Commits
Show all changes
7 commits
Select commit
Hold shift + click to select a range
79e3c87
Add prompt-caching examples for Effect AI
subtleGradient d53c86d
Run biome format
subtleGradient 5219b44
Simplify Effect AI prompt-caching README to link to main docs
subtleGradient cabf912
Rename prompt caching examples with anthropic prefix
subtleGradient 3305d85
Use glob pattern runner for effect-ai examples
subtleGradient af2da62
Remove filename references from docs to prevent sync issues
subtleGradient a168119
docs: link to OpenRouter docs and remove duplicated content
subtleGradient File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,48 @@ | ||
| # Effect-TS AI Examples | ||
|
|
||
| Examples using Effect-TS with @effect/ai and @effect/ai-openrouter for type-safe, composable AI operations. | ||
|
|
||
| ## Prerequisites | ||
|
|
||
| - Bun runtime: `curl -fsSL https://bun.sh/install | bash` | ||
| - `OPENROUTER_API_KEY` environment variable | ||
|
|
||
| ## Running Examples | ||
|
|
||
| ```bash | ||
| # From monorepo root (typescript/) | ||
| bun examples | ||
|
|
||
| # Or from this workspace | ||
| cd effect-ai | ||
| bun examples | ||
| ``` | ||
|
|
||
| ## Features | ||
|
|
||
| - [prompt-caching](./src/prompt-caching/) - Anthropic caching examples with Effect patterns | ||
|
|
||
| ### Key Configuration | ||
|
|
||
| **CRITICAL**: The Effect AI example requires: | ||
| ```typescript | ||
| config: { | ||
| stream_options: { include_usage: true } | ||
| } | ||
| ``` | ||
|
|
||
| Without this, `usage.cachedInputTokens` will be undefined in the response. | ||
|
|
||
| ### Effect Patterns Demonstrated | ||
|
|
||
| - `Effect.gen` for generator-based composition | ||
| - Layer-based dependency injection | ||
| - Type-safe error handling | ||
| - Evidence-based validation | ||
|
|
||
| ## Dependencies | ||
|
|
||
| - `@openrouter-examples/shared` - Shared constants (LARGE_SYSTEM_PROMPT) and types | ||
| - `@effect/ai` - Effect AI abstractions | ||
| - `@effect/ai-openrouter` - OpenRouter provider for Effect AI | ||
| - `effect` - Effect-TS core library |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,21 @@ | ||
| { | ||
| "name": "@openrouter-examples/effect-ai", | ||
| "version": "1.0.0", | ||
| "private": true, | ||
| "type": "module", | ||
| "scripts": { | ||
| "examples": "bun run run-examples.ts", | ||
| "typecheck": "tsc --noEmit" | ||
| }, | ||
| "dependencies": { | ||
| "@openrouter-examples/shared": "workspace:*", | ||
| "@effect/ai": "^0.32.1", | ||
| "@effect/ai-openrouter": "^0.6.0", | ||
| "@effect/platform": "^0.93.0", | ||
| "@effect/platform-bun": "^0.83.0", | ||
| "effect": "^3.19.3" | ||
| }, | ||
| "devDependencies": { | ||
| "@types/bun": "latest" | ||
| } | ||
| } | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,57 @@ | ||
| #!/usr/bin/env bun | ||
| /** | ||
| * Run all example files in the src/ directory | ||
| * Each example is run in a separate process to handle process.exit() calls | ||
| */ | ||
|
|
||
| import { readdirSync, statSync } from 'fs'; | ||
| import { join } from 'path'; | ||
| import { $ } from 'bun'; | ||
|
|
||
| const srcDir = join(import.meta.dir, 'src'); | ||
|
|
||
| // Recursively find all .ts files in src/ | ||
| function findExamples(dir: string): string[] { | ||
| const entries = readdirSync(dir); | ||
| const files: string[] = []; | ||
|
|
||
| for (const entry of entries) { | ||
| const fullPath = join(dir, entry); | ||
| const stat = statSync(fullPath); | ||
|
|
||
| if (stat.isDirectory()) { | ||
| files.push(...findExamples(fullPath)); | ||
| } else if (entry.endsWith('.ts')) { | ||
| files.push(fullPath); | ||
| } | ||
| } | ||
|
|
||
| return files.sort(); | ||
| } | ||
|
|
||
| const examples = findExamples(srcDir); | ||
| console.log(`Found ${examples.length} example(s)\n`); | ||
|
|
||
| let failed = 0; | ||
| for (const example of examples) { | ||
| const relativePath = example.replace(import.meta.dir + '/', ''); | ||
| console.log(`\n${'='.repeat(80)}`); | ||
| console.log(`Running: ${relativePath}`); | ||
| console.log('='.repeat(80)); | ||
|
|
||
| try { | ||
| await $`bun run ${example}`.quiet(); | ||
| console.log(`✅ ${relativePath} completed successfully`); | ||
| } catch (error) { | ||
| console.error(`❌ ${relativePath} failed`); | ||
| failed++; | ||
| } | ||
| } | ||
|
|
||
| console.log(`\n${'='.repeat(80)}`); | ||
| console.log(`Results: ${examples.length - failed}/${examples.length} passed`); | ||
| console.log('='.repeat(80)); | ||
|
|
||
| if (failed > 0) { | ||
| process.exit(1); | ||
| } |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,12 @@ | ||
| # Prompt Caching Examples (Effect AI) | ||
|
|
||
| Examples demonstrating prompt caching with @effect/ai and @effect/ai-openrouter. | ||
|
|
||
| ## Documentation | ||
|
|
||
| For full prompt caching documentation including all providers, pricing, and configuration details, see: | ||
| - **[OpenRouter Prompt Caching Guide](https://openrouter.ai/docs/features/prompt-caching)** | ||
|
|
||
| ## Examples in This Directory | ||
|
|
||
| See the TypeScript files in this directory for specific examples. |
119 changes: 119 additions & 0 deletions
119
typescript/effect-ai/src/prompt-caching/anthropic-multi-message-cache.ts
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,119 @@ | ||
| /** | ||
| * Example: Anthropic Prompt Caching - Multi-Message Conversation (Effect AI) | ||
| * | ||
| * This example demonstrates Anthropic prompt caching in a multi-message conversation | ||
| * via OpenRouter using Effect AI. | ||
| * | ||
| * Pattern: User message cache in multi-turn conversation using Effect patterns | ||
| */ | ||
|
|
||
| import * as OpenRouterClient from '@effect/ai-openrouter/OpenRouterClient'; | ||
| import * as OpenRouterLanguageModel from '@effect/ai-openrouter/OpenRouterLanguageModel'; | ||
| import * as LanguageModel from '@effect/ai/LanguageModel'; | ||
| import * as Prompt from '@effect/ai/Prompt'; | ||
| import { FetchHttpClient } from '@effect/platform'; | ||
| import * as BunContext from '@effect/platform-bun/BunContext'; | ||
| import { LARGE_SYSTEM_PROMPT } from '@openrouter-examples/shared/constants'; | ||
| import { Console, Effect, Layer, Redacted } from 'effect'; | ||
|
|
||
| const program = Effect.gen(function* () { | ||
| const testId = Date.now(); | ||
| const largeContext = `Test ${testId}: Context:\n\n${LARGE_SYSTEM_PROMPT}`; | ||
|
|
||
| yield* Console.log( | ||
| '╔════════════════════════════════════════════════════════════════════════════╗', | ||
| ); | ||
| yield* Console.log( | ||
| '║ Anthropic Prompt Caching - Multi-Message (Effect AI) ║', | ||
| ); | ||
| yield* Console.log( | ||
| '╚════════════════════════════════════════════════════════════════════════════╝', | ||
| ); | ||
| yield* Console.log(''); | ||
| yield* Console.log('Testing cache_control in multi-turn conversation'); | ||
| yield* Console.log(''); | ||
|
|
||
| const makePrompt = () => | ||
| Prompt.make([ | ||
| { | ||
| role: 'user' as const, | ||
| content: [ | ||
| { | ||
| type: 'text' as const, | ||
| text: largeContext, | ||
| options: { | ||
| openrouter: { | ||
| cacheControl: { type: 'ephemeral' as const }, | ||
| }, | ||
| }, | ||
| }, | ||
| { | ||
| type: 'text' as const, | ||
| text: "Hello, what's your purpose?", | ||
| }, | ||
| ], | ||
| }, | ||
| { | ||
| role: 'assistant' as const, | ||
| content: "I'm an AI assistant designed to help with various tasks.", | ||
| }, | ||
| { | ||
| role: 'user' as const, | ||
| content: 'What programming languages do you know?', | ||
| }, | ||
| ]); | ||
|
|
||
| yield* Console.log('First Call (Cache Miss Expected)'); | ||
| const response1 = yield* LanguageModel.generateText({ | ||
| prompt: makePrompt(), | ||
| }); | ||
| const cached1 = response1.usage.cachedInputTokens ?? 0; | ||
| yield* Console.log(` Response: ${response1.text.substring(0, 80)}...`); | ||
| yield* Console.log(` cached_tokens=${cached1}`); | ||
|
|
||
| yield* Effect.sleep('1 second'); | ||
|
|
||
| yield* Console.log('\nSecond Call (Cache Hit Expected)'); | ||
| const response2 = yield* LanguageModel.generateText({ | ||
| prompt: makePrompt(), | ||
| }); | ||
| const cached2 = response2.usage.cachedInputTokens ?? 0; | ||
| yield* Console.log(` Response: ${response2.text.substring(0, 80)}...`); | ||
| yield* Console.log(` cached_tokens=${cached2}`); | ||
|
|
||
| // Analysis | ||
| yield* Console.log('\n' + '='.repeat(80)); | ||
| yield* Console.log('ANALYSIS'); | ||
| yield* Console.log('='.repeat(80)); | ||
| yield* Console.log(`First call: cached_tokens=${cached1} (expected: 0)`); | ||
| yield* Console.log(`Second call: cached_tokens=${cached2} (expected: >0)`); | ||
|
|
||
| const success = cached1 === 0 && cached2 > 0; | ||
|
|
||
| if (success) { | ||
| yield* Console.log('\n✓ SUCCESS - Multi-message caching is working correctly'); | ||
| } else { | ||
| yield* Console.log('\n✗ FAILURE - Multi-message caching is not working as expected'); | ||
| } | ||
|
|
||
| yield* Console.log('='.repeat(80)); | ||
| }); | ||
|
|
||
| const OpenRouterClientLayer = OpenRouterClient.layer({ | ||
| apiKey: Redacted.make(process.env.OPENROUTER_API_KEY!), | ||
| }).pipe(Layer.provide(FetchHttpClient.layer)); | ||
|
|
||
| const OpenRouterModelLayer = OpenRouterLanguageModel.layer({ | ||
| model: 'anthropic/claude-3.5-sonnet', | ||
| config: { | ||
| stream_options: { include_usage: true }, | ||
| }, | ||
| }).pipe(Layer.provide(OpenRouterClientLayer)); | ||
|
|
||
| await program.pipe( | ||
| Effect.provide(OpenRouterModelLayer), | ||
| Effect.provide(BunContext.layer), | ||
| Effect.runPromise, | ||
| ); | ||
|
|
||
| console.log('\n✓ Program completed successfully'); |
111 changes: 111 additions & 0 deletions
111
typescript/effect-ai/src/prompt-caching/anthropic-no-cache-control.ts
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,111 @@ | ||
| /** | ||
| * Example: Anthropic Prompt Caching - Control (No cache_control) (Effect AI) | ||
| * | ||
| * This is a CONTROL scenario demonstrating that without cache_control, | ||
| * no caching occurs. | ||
| * | ||
| * Purpose: Validates that cache behavior is due to cache_control, not coincidence | ||
| */ | ||
|
|
||
| import * as OpenRouterClient from '@effect/ai-openrouter/OpenRouterClient'; | ||
| import * as OpenRouterLanguageModel from '@effect/ai-openrouter/OpenRouterLanguageModel'; | ||
| import * as LanguageModel from '@effect/ai/LanguageModel'; | ||
| import * as Prompt from '@effect/ai/Prompt'; | ||
| import { FetchHttpClient } from '@effect/platform'; | ||
| import * as BunContext from '@effect/platform-bun/BunContext'; | ||
| import { LARGE_SYSTEM_PROMPT } from '@openrouter-examples/shared/constants'; | ||
| import { Console, Effect, Layer, Redacted } from 'effect'; | ||
|
|
||
| const program = Effect.gen(function* () { | ||
| const testId = Date.now(); | ||
| const largeContext = `Test ${testId}: Context:\n\n${LARGE_SYSTEM_PROMPT}`; | ||
|
|
||
| yield* Console.log( | ||
| '╔════════════════════════════════════════════════════════════════════════════╗', | ||
| ); | ||
| yield* Console.log( | ||
| '║ Anthropic Prompt Caching - Control (No cache_control) (Effect AI) ║', | ||
| ); | ||
| yield* Console.log( | ||
| '╚════════════════════════════════════════════════════════════════════════════╝', | ||
| ); | ||
| yield* Console.log(''); | ||
| yield* Console.log('Testing WITHOUT cache_control (control scenario)'); | ||
| yield* Console.log(''); | ||
|
|
||
| const makePrompt = () => | ||
| Prompt.make([ | ||
| { | ||
| role: 'user' as const, | ||
| content: [ | ||
| { | ||
| type: 'text' as const, | ||
| text: largeContext, | ||
| // NO cache_control - this is the control | ||
| }, | ||
| { | ||
| type: 'text' as const, | ||
| text: 'What are the key principles?', | ||
| }, | ||
| ], | ||
| }, | ||
| ]); | ||
|
Comment on lines
+36
to
+52
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Would pull this into a separate function |
||
|
|
||
| yield* Console.log('First Call (No Cache Expected)'); | ||
| const response1 = yield* LanguageModel.generateText({ | ||
| prompt: makePrompt(), | ||
| }); | ||
| const cached1 = response1.usage.cachedInputTokens ?? 0; | ||
| yield* Console.log(` cached_tokens=${cached1}`); | ||
|
|
||
| yield* Effect.sleep('1 second'); | ||
|
|
||
| yield* Console.log('\nSecond Call (No Cache Expected)'); | ||
| const response2 = yield* LanguageModel.generateText({ | ||
| prompt: makePrompt(), | ||
| }); | ||
| const cached2 = response2.usage.cachedInputTokens ?? 0; | ||
| yield* Console.log(` cached_tokens=${cached2}`); | ||
|
|
||
| // Analysis | ||
| yield* Console.log('\n' + '='.repeat(80)); | ||
| yield* Console.log('ANALYSIS (CONTROL)'); | ||
| yield* Console.log('='.repeat(80)); | ||
| yield* Console.log(`First call: cached_tokens=${cached1} (expected: 0)`); | ||
| yield* Console.log(`Second call: cached_tokens=${cached2} (expected: 0)`); | ||
|
|
||
| if (cached1 === 0 && cached2 === 0) { | ||
| yield* Console.log('✓ No cache metrics present (expected for control)'); | ||
| } else { | ||
| yield* Console.log('✗ Unexpected cache metrics in control scenario'); | ||
| } | ||
|
|
||
| const success = cached1 === 0 && cached2 === 0; | ||
|
|
||
| if (success) { | ||
| yield* Console.log('\n✓ SUCCESS - Control scenario confirms no false positives'); | ||
| } else { | ||
| yield* Console.log('\n✗ FAILURE - Control scenario shows unexpected cache behavior'); | ||
| } | ||
|
|
||
| yield* Console.log('='.repeat(80)); | ||
| }); | ||
|
|
||
| const OpenRouterClientLayer = OpenRouterClient.layer({ | ||
| apiKey: Redacted.make(process.env.OPENROUTER_API_KEY!), | ||
| }).pipe(Layer.provide(FetchHttpClient.layer)); | ||
|
|
||
| const OpenRouterModelLayer = OpenRouterLanguageModel.layer({ | ||
| model: 'anthropic/claude-3.5-sonnet', | ||
| config: { | ||
| stream_options: { include_usage: true }, | ||
| }, | ||
| }).pipe(Layer.provide(OpenRouterClientLayer)); | ||
|
|
||
| await program.pipe( | ||
| Effect.provide(OpenRouterModelLayer), | ||
| Effect.provide(BunContext.layer), | ||
| Effect.runPromise, | ||
| ); | ||
|
|
||
| console.log('\n✓ Program completed successfully'); | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
freeze them deps