llm-unified 0.1.3

Unified LLM provider layer: one trait, many backends (OpenAI, Anthropic, DeepSeek, Qwen, ...)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

coverage:
  status:
    project:
      default:
        target: 85%
        threshold: 2%
    patch:
      default:
        # Per-diff bar: new code should be covered, but small glue diffs should
        # not be blocked by a single uncovered line.
        target: 80%
        threshold: 5%

ignore:
  - "tests/**"
  - "examples/**"
  - "fuzz/**"
  # Binary entry point: argument parsing + printing, exercised manually via
  # `cargo run --bin llm-cli`. Excluded so the number reflects library logic.
  - "src/bin/**"

comment:
  # One coverage comment per PR, updated in place.
  behavior: default
  require_changes: false