contribute-now 0.9.0-dev.30c88ec → 0.9.0-dev.51afb57

package/README.md

@@ -324,27 +324,6 @@ cn label suggest --issue 42
 
 # Get ranked label suggestions for a PR
 cn label suggest --pr 7
-
-# Auto-apply top labels to a PR
-cn label apply --pr 7
-
-# Auto-apply top labels to an issue
-cn label apply --issue 42
-
-# Bulk preview (safe default): inspect open issues and PRs
-cn label apply
-
-# Bulk apply to open issues and PRs
-cn label apply --yes
-
-# Bulk apply to PRs only with custom limits
-cn label apply --prs --yes --limit 30 --count 2 --min-score 5
-
-# Force heuristic-only ranking (no AI)
-cn label apply --pr 7 --no-ai
-
-# Use a specific AI model for ranking
-cn label apply --pr 7 --model gpt-4.1
 ```
 
 **Label source strategy:**
@@ -354,15 +333,6 @@ cn label apply --pr 7 --model gpt-4.1
 4. The local cache is used by default — no repeated `gh` API calls.
 5. On label-not-found errors, the cache is automatically resynced and the operation is retried once.
 
-**Auto-apply behavior (`cn label apply`):**
-- Uses the same label scoring engine as `cn label suggest` (label names + descriptions vs issue/PR content).
-- Applies only existing repository labels (never creates labels).
-- Filters out labels that are already present on the target issue/PR.
-- In bulk mode (no `--issue`/`--pr`), defaults to dry-run unless `--yes` is provided.
-- Supports tunable controls: `--count`, `--min-score`, `--limit`, `--issues`, `--prs`, and `--dry-run`.
-- Uses AI ranking by default when AI is enabled in your config, with automatic fallback to heuristic scoring.
-- Pass `--no-ai` to force heuristic-only scoring or `--model <name>` to pick a specific AI model.
-
 **Label input format:**
 - Commas are the separator between labels.
 - Spaces are part of a label name (`good first issue` is one label, not three words).