agnes-the-ai-analyst

AI-Cognitive-Leap/agnes-the-ai-analyst

Fork 0

Commit graph

Author	SHA1	Message	Date
Petr	c04791b702	Suppress httpcore debug logging in LLM connector	2026-03-23 12:57:35 +01:00
Petr	f619fadc42	Fix SSL verification and suppress OpenAI SDK debug logging - Add verify_ssl config option for corporate proxies with self-signed certs - Suppress openai/httpx debug loggers that dump full request bodies (including prompt content) — security requirement	2026-03-23 12:56:04 +01:00
Petr	95358448e6	Add modular LLM connector for Corporate Memory Replace hardwired Anthropic API calls with a pluggable provider system. Each deployment configures its AI provider in instance.yaml — switching between Anthropic, LiteLLM, OpenRouter, or any OpenAI-compatible proxy is a config change, not a code change. New connectors/llm/ module: - StructuredExtractor Protocol with extract_json() interface - AnthropicExtractor: direct Anthropic SDK with retry + backoff - OpenAICompatExtractor: any OpenAI-compatible proxy with three-layer structured output fallback (json_schema -> json_object -> prompt) - Configurable structured_output policy (strict/json/auto) - Custom exception hierarchy (auth/rate_limit/timeout/format/refusal) - Zero secrets in logs: no API keys, prompts, or responses logged Reviewed by: Google Gemini, Claude Sonnet, OpenAI GPT-5.4. Security audit passed with all critical findings resolved.	2026-03-23 12:08:33 +01:00

Author

SHA1

Message

Date

Petr

c04791b702

Suppress httpcore debug logging in LLM connector

2026-03-23 12:57:35 +01:00

Petr

f619fadc42

Fix SSL verification and suppress OpenAI SDK debug logging

- Add verify_ssl config option for corporate proxies with self-signed certs
- Suppress openai/httpx debug loggers that dump full request bodies
  (including prompt content) — security requirement

2026-03-23 12:56:04 +01:00

Petr

95358448e6

Add modular LLM connector for Corporate Memory

Replace hardwired Anthropic API calls with a pluggable provider system.
Each deployment configures its AI provider in instance.yaml — switching
between Anthropic, LiteLLM, OpenRouter, or any OpenAI-compatible proxy
is a config change, not a code change.

New connectors/llm/ module:
- StructuredExtractor Protocol with extract_json() interface
- AnthropicExtractor: direct Anthropic SDK with retry + backoff
- OpenAICompatExtractor: any OpenAI-compatible proxy with three-layer
  structured output fallback (json_schema -> json_object -> prompt)
- Configurable structured_output policy (strict/json/auto)
- Custom exception hierarchy (auth/rate_limit/timeout/format/refusal)
- Zero secrets in logs: no API keys, prompts, or responses logged

Reviewed by: Google Gemini, Claude Sonnet, OpenAI GPT-5.4.
Security audit passed with all critical findings resolved.

2026-03-23 12:08:33 +01:00

3 commits