docs: add asqav signed audit trail callback integration

2026-09-16 23:41:43 +00:00 · 2026-04-08 10:24:06 +02:00 · 2026-04-08 10:24:06 +02:00 · 2b1594b860
commit 2b1594b860
parent b8f7d61400
1 changed files with 126 additions and 0 deletions
--- a/docs/my-website/docs/observability/asqav_integration.md
+++ b/docs/my-website/docs/observability/asqav_integration.md
@ -0,0 +1,126 @@
+import Tabs from '@theme/Tabs';
+import TabItem from '@theme/TabItem';
+
+# Asqav - Signed Audit Trail & AI Governance
+
+[Asqav](https://asqav.com/) provides cryptographically signed audit trails and governance for AI applications. Every LLM call is logged with a tamper-evident signature, giving compliance teams verifiable records of model inputs, outputs, and metadata.
+
+**Key Features:**
+
+- **Signed audit logs** - Every call produces a cryptographically signed record, verifiable offline
+- **Governance callbacks** - Hook into LiteLLM's callback system with zero changes to your inference code
+- **Compliance-ready exports** - Structured logs compatible with SOC 2, ISO 27001, and EU AI Act requirements
+- **Model and cost tracking** - Token usage, latency, and cost captured per request
+
+:::tip
+
+This is community maintained. Please make an issue if you run into a bug:
+https://github.com/BerriAI/litellm
+
+:::
+
+## Pre-Requisites
+
+```bash
+pip install "asqav[litellm]"
+```
+
+Get your API key from [app.asqav.com](https://app.asqav.com/).
+
+## Quick Start
+
+<Tabs>
+<TabItem value="sdk" label="Python SDK">
+
+```python
+import os
+import litellm
+from asqav.extras.litellm import AsqavGuardrail
+
+os.environ["ASQAV_API_KEY"] = "your-asqav-api-key"
+os.environ["OPENAI_API_KEY"] = "your-openai-api-key"
+
+# Register the Asqav callback
+asqav_handler = AsqavGuardrail()
+litellm.callbacks = [asqav_handler]
+
+response = litellm.completion(
+    model="gpt-4o-mini",
+    messages=[{"role": "user", "content": "Summarize the EU AI Act."}],
+)
+
+print(response)
+```
+
+Every call is logged to Asqav with a signed audit record. No other changes to your code are needed.
+
+</TabItem>
+<TabItem value="proxy" label="LiteLLM Proxy">
+
+### 1. Install the package on the proxy server
+
+```bash
+pip install "asqav[litellm]"
+```
+
+### 2. Add Asqav to your `config.yaml`
+
+```yaml
+model_list:
+  - model_name: gpt-4o-mini
+    litellm_params:
+      model: openai/gpt-4o-mini
+      api_key: os.environ/OPENAI_API_KEY
+
+litellm_settings:
+  callbacks: ["asqav.extras.litellm.AsqavGuardrail"]
+
+general_settings:
+  master_key: "sk-1234"
+
+environment_variables:
+  ASQAV_API_KEY: "your-asqav-api-key"
+```
+
+### 3. Start the proxy
+
+```bash
+litellm --config config.yaml
+```
+
+### 4. Test it
+
+```bash
+curl -X POST 'http://0.0.0.0:4000/chat/completions' \
+  -H 'Content-Type: application/json' \
+  -H 'Authorization: Bearer sk-1234' \
+  -d '{"model": "gpt-4o-mini", "messages": [{"role": "user", "content": "Hello"}]}'
+```
+
+Signed audit records appear immediately in your [Asqav dashboard](https://app.asqav.com/).
+
+</TabItem>
+</Tabs>
+
+## Environment Variables
+
+| Variable         | Description                                  |
+| ---------------- | -------------------------------------------- |
+| `ASQAV_API_KEY`  | Your Asqav API key for authentication        |
+
+## What Gets Logged?
+
+The [LiteLLM Standard Logging Payload](https://docs.litellm.ai/docs/proxy/logging_spec) is captured on each successful LLM API call, including:
+
+- Request messages and model parameters
+- Response content and finish reason
+- Token usage and cost
+- Latency metrics
+- A cryptographic signature over the full record
+
+Records are retrievable and verifiable via the Asqav API or dashboard.
+
+## Support
+
+- Docs: [docs.asqav.com](https://docs.asqav.com/)
+- Issues: [github.com/asqav/asqav-sdk](https://github.com/asqav/asqav-sdk/issues)