@peac/adapter-openai-compatible 0.12.5 → 0.12.7

package/README.md

@@ -1,23 +1,24 @@
 # @peac/adapter-openai-compatible
 
-OpenAI-compatible chat completion adapter for PEAC interaction evidence.
+OpenAI-compatible chat completion adapter for PEAC interaction evidence using a hash-first model.
 
-Maps OpenAI-compatible chat completion responses into PEAC `InteractionEvidenceV01` using a **hash-first model** (DD-138): no raw prompt or completion text is stored in receipts. Only SHA-256 digests, model identifiers, token counts, and timing metadata are recorded.
-
-Works with any OpenAI-compatible provider (OpenAI, Anthropic Messages API via adapter, Ollama, vLLM, Together, etc.) without importing their SDKs.
-
-## Install
+## Installation
 
 ```bash
 pnpm add @peac/adapter-openai-compatible
 ```
 
-## Usage
+## What It Does
+
+`@peac/adapter-openai-compatible` maps chat completion responses from any OpenAI-compatible provider into PEAC interaction evidence. It uses a hash-first model: only SHA-256 digests of messages and output are recorded, along with model identifiers, token counts, and timing metadata. No raw prompt or completion text ever appears in the evidence structure. Works with OpenAI, Anthropic (via adapter), Ollama, vLLM, Together, and any other OpenAI-compatible provider without importing their SDKs.
+
+## How Do I Use It?
+
+### Create evidence from a chat completion
 
 ```typescript
 import { fromChatCompletion } from '@peac/adapter-openai-compatible';
 
-// After making an OpenAI-compatible API call:
 const evidence = await fromChatCompletion({
   messages: [
     { role: 'system', content: 'You are a helpful assistant.' },
@@ -37,45 +38,47 @@ const evidence = await fromChatCompletion({
     ],
     usage: { prompt_tokens: 20, completion_tokens: 5, total_tokens: 25 },
   },
-  provider: 'openai', // optional
+  provider: 'openai',
 });
 
-// evidence.input.digest  -> SHA-256 of messages (no raw text)
-// evidence.output.digest -> SHA-256 of output (no raw text)
+// evidence.input.digest  -> SHA-256 of canonicalized messages (no raw text)
+// evidence.output.digest -> SHA-256 of output content (no raw text)
 // evidence.extensions    -> model, usage, finish_reason
 ```
 
-## Hash-First Model
-
-Per DD-138, this adapter records only:
+### Hash messages and output independently
 
-| Field           | Content                                 |
-| --------------- | --------------------------------------- |
-| `input.digest`  | SHA-256 of canonicalized messages array |
-| `output.digest` | SHA-256 of concatenated output content  |
-| `executor`      | Platform identifier and model name      |
-| `extensions`    | Model ID, token counts, finish reason   |
+```typescript
+import { hashMessages, hashOutput } from '@peac/adapter-openai-compatible';
 
-Raw prompt and completion text never appear in the evidence structure.
+const inputHash = await hashMessages([{ role: 'user', content: 'Hello' }]);
+// 'sha256:...'
 
-## Streaming
+const outputHash = await hashOutput('The answer is 4.');
+// 'sha256:...'
+```
 
-Streaming support (`fromChatCompletionStream`) is explicitly deferred to v0.11.3.
+## Integrates With
 
-## API
+- `@peac/kernel` (Layer 0): Wire constants and types
+- `@peac/schema` (Layer 1): Interaction evidence schemas
+- `@peac/protocol` (Layer 3): Receipt issuance with mapped inference evidence
 
-### `fromChatCompletion(params): Promise<InferenceEvidence>`
+## For Agent Developers
 
-Maps a chat completion response to interaction evidence.
+If you are building an AI agent or service that calls OpenAI-compatible LLM APIs:
 
-### `hashMessages(messages): Promise<string>`
+- Use `fromChatCompletion()` after each API call to produce signed evidence of the interaction
+- The hash-first model ensures no prompt or completion text leaks into receipts
+- The adapter is provider-agnostic; pass `provider` to tag evidence with the source platform
+- See the [llms.txt](https://github.com/peacprotocol/peac/blob/main/llms.txt) for a concise protocol overview
 
-SHA-256 hash of a messages array. Returns `sha256:<hex64>`.
+## License
 
-### `hashOutput(content): Promise<string>`
+Apache-2.0
 
-SHA-256 hash of output text. Returns `sha256:<hex64>`.
+---
 
-## License
+PEAC Protocol is an open source project stewarded by Originary and community contributors.
 
-Apache-2.0
+[Docs](https://www.peacprotocol.org) | [GitHub](https://github.com/peacprotocol/peac) | [Originary](https://www.originary.xyz)

package/package.json

@@ -1,15 +1,15 @@
 {
   "name": "@peac/adapter-openai-compatible",
-  "version": "0.12.5",
+  "version": "0.12.7",
   "description": "OpenAI-compatible chat completion adapter for PEAC interaction evidence (hash-first)",
   "main": "dist/index.cjs",
   "types": "dist/index.d.ts",
   "repository": {
     "type": "git",
-    "url": "https://github.com/peacprotocol/peac.git",
+    "url": "git+https://github.com/peacprotocol/peac.git",
     "directory": "packages/adapters/openai-compatible"
   },
-  "author": "jithinraj <7850727+jithinraj@users.noreply.github.com>",
+  "author": "PEAC Protocol Contributors",
   "license": "Apache-2.0",
   "bugs": {
     "url": "https://github.com/peacprotocol/peac/issues"
@@ -23,8 +23,8 @@
     "access": "public"
   },
   "dependencies": {
-    "@peac/kernel": "0.12.5",
-    "@peac/schema": "0.12.5"
+    "@peac/kernel": "0.12.7",
+    "@peac/schema": "0.12.7"
   },
   "devDependencies": {
     "@types/node": "^22.19.11",
@@ -41,6 +41,18 @@
     },
     "./package.json": "./package.json"
   },
+  "keywords": [
+    "peac",
+    "peacprotocol",
+    "interaction-records",
+    "signed-records",
+    "receipts",
+    "originary",
+    "openai-compatible",
+    "inference",
+    "response-evidence",
+    "llm"
+  ],
   "scripts": {
     "prebuild": "rm -rf dist",
     "build": "pnpm run build:js && pnpm run build:types",