feat: add --show-after output filter to classify

README.md

@@ -114,6 +114,11 @@ uv run review-classify classify --repo owner/repo --format json > outliers.json
 
 # Exclude PRs merged into the primary branch (main/master)
 uv run review-classify classify --repo owner/repo --exclude-primary-merged
+
+# Show only outliers merged in the last two weeks (stats baseline unchanged)
+uv run review-classify classify --repo owner/repo \
+  --start 2024-01-01 --end 2024-11-30 \
+  --show-after 2024-12-15
 ```
 
 | Option | Description |
@@ -127,6 +132,7 @@ uv run review-classify classify --repo owner/repo --exclude-primary-merged
 | `--start` | Start of the classification window (YYYY-MM-DD). |
 | `--end` | End of the classification window (YYYY-MM-DD). |
 | `--exclude-primary-merged` | Exclude PRs whose base branch is `main` or `master`. |
+| `--show-after` | Only display outliers merged after this date (YYYY-MM-DD). Stats baseline is unaffected. |
 | `--verbose` / `-v` | Print progress details. |
 
 #### Classification window (`--start` / `--end`)
@@ -163,6 +169,26 @@ Pass `--exclude-primary-merged` to restrict analysis to PRs that were **not** me
 uv run review-classify classify --repo owner/repo --exclude-primary-merged
 ```
 
+#### Filtering output to recent PRs (`--show-after`)
+
+`--show-after YYYY-MM-DD` filters the **displayed** outliers to only those merged after the given date. The stats baseline (`--start` / `--end`) is computed from the full window as normal — only the output is narrowed.
+
+This is the right tool when you have a long history in the database but only want to act on recent outliers. For example, compute stats from a 6-month baseline but display only the last two weeks:
+
+```
+[--start ──────────────────── --end]   >end … [--show-after] … today
+      ↑                          ↑                    ↑
+ baseline start            baseline end        only show from here
+```
+
+```bash
+# Compute stats from Jan–Nov 2024; display only outliers merged after 2024-12-15
+uv run review-classify classify --repo owner/repo \
+  --start 2024-01-01 \
+  --end   2024-11-30 \
+  --show-after 2024-12-15
+```
+
 ### Per-repository analysis
 
 Outlier detection is always **scoped to a single repository**. When you target multiple repositories (via `--org`, multiple `--repo` flags, or a config file), each repository is analysed independently:
@@ -222,6 +248,7 @@ threshold     = 2.0
 min_samples   = 30
 start         = "2024-01-01"
 end           = "2024-06-30"
+show_after    = "2024-12-15"   # only display outliers merged after this date
 
 # Individual repositories ─────────────────────────────────────────────────────
 

src/review_classification/cli/app.py

@@ -103,13 +103,15 @@ def _detect_single(
     classify_start: str | None,
     classify_end: str | None,
     exclude_primary_merged: bool = False,
+    show_after: str | None = None,
 ) -> RepoClassifyResult:
     """Run outlier detection for a single repository."""
     repo = GitHubRepo.from_string(repo_name)
     full_name = f"{repo.owner}/{repo.name}"
 
     classify_start_dt: datetime | None = None
     classify_end_dt: datetime | None = None
+    show_after_dt: datetime | None = None
 
     if classify_start:
         classify_start_dt = datetime.strptime(classify_start, "%Y-%m-%d").replace(
@@ -119,6 +121,8 @@ def _detect_single(
         classify_end_dt = datetime.strptime(classify_end, "%Y-%m-%d").replace(
             hour=23, minute=59, second=59, tzinfo=UTC
         )
+    if show_after:
+        show_after_dt = datetime.strptime(show_after, "%Y-%m-%d").replace(tzinfo=UTC)
 
     exclude_branches = _PRIMARY_BRANCHES if exclude_primary_merged else None
 
@@ -130,6 +134,8 @@ def _detect_single(
             start_label = classify_start or "unbounded"
             end_label = classify_end or "unbounded"
             typer.echo(f"Classification window: {start_label} to {end_label}")
+        if show_after_dt:
+            typer.echo(f"Showing outliers merged after: {show_after}")
         if exclude_primary_merged:
             branches = sorted(_PRIMARY_BRANCHES)
             typer.echo(f"Excluding PRs merged into primary branches: {branches}")
@@ -192,6 +198,11 @@ def _detect_single(
 
         save_outlier_scores(session, full_name, results, sample_size)
 
+        if show_after_dt:
+            results = [
+                r for r in results if r.merged_at and r.merged_at > show_after_dt
+            ]
+
         outliers = [r for r in results if r.is_outlier]
 
         if verbose:
@@ -243,6 +254,7 @@ def _resolve_targets(
     default_min_samples: int | None = None,
     default_start: str | None = None,
     default_end: str | None = None,
+    default_show_after: str | None = None,
 ) -> list[RepoConfig]:
     """Resolve CLI arguments and config file into a concrete list of repositories."""
     from ..queries.github_client import get_org_repos
@@ -293,6 +305,11 @@ def add_repo(rc: RepoConfig) -> None:
                             org_cfg.start if org_cfg.start is not None else multi.start
                         ),
                         end=(org_cfg.end if org_cfg.end is not None else multi.end),
+                        show_after=(
+                            org_cfg.show_after
+                            if org_cfg.show_after is not None
+                            else multi.show_after
+                        ),
                     )
                     add_repo(rc)
 
@@ -306,6 +323,7 @@ def add_repo(rc: RepoConfig) -> None:
             min_samples=default_min_samples,
             start=default_start,
             end=default_end,
+            show_after=default_show_after,
         )
         add_repo(rc)
 
@@ -321,6 +339,7 @@ def add_repo(rc: RepoConfig) -> None:
                 min_samples=default_min_samples,
                 start=default_start,
                 end=default_end,
+                show_after=default_show_after,
             )
             add_repo(rc)
 
@@ -481,6 +500,16 @@ def classify(
             help="Exclude PRs merged into the primary branch (main/master).",
         ),
     ] = False,
+    show_after: Annotated[
+        str | None,
+        typer.Option(
+            "--show-after",
+            help=(
+                "Only display outliers merged after this date (YYYY-MM-DD). "
+                "Stats baseline is unaffected — use this to focus output on recent PRs."
+            ),
+        ),
+    ] = None,
 ) -> None:
     """Classify previously fetched PRs using z-score outlier analysis.
 
@@ -513,6 +542,7 @@ def classify(
             default_min_samples=min_samples,
             default_start=start,
             default_end=end,
+            default_show_after=show_after,
         )
 
         repo_results = [
@@ -524,6 +554,7 @@ def classify(
                 target.start,
                 target.end,
                 exclude_primary_merged=exclude_primary_merged,
+                show_after=target.show_after,
             )
             for target in targets
         ]

src/review_classification/cli/config.py

@@ -16,6 +16,7 @@ class RepoConfig:
     min_samples: int | None = None
     start: str | None = None
     end: str | None = None
+    show_after: str | None = None
 
 
 @dataclass
@@ -29,6 +30,7 @@ class OrgConfig:
     min_samples: int | None = None
     start: str | None = None
     end: str | None = None
+    show_after: str | None = None
     exclude_repos: list[str] = field(default_factory=list)
 
 
@@ -49,6 +51,7 @@ class MultiRepoConfig:
     min_samples: int = 30
     start: str | None = None
     end: str | None = None
+    show_after: str | None = None
 
     def resolve(self, repo: RepoConfig) -> RepoConfig:
         """Return a RepoConfig with global defaults applied for unset fields.
@@ -77,6 +80,9 @@ def resolve(self, repo: RepoConfig) -> RepoConfig:
             ),
             start=repo.start if repo.start is not None else self.start,
             end=repo.end if repo.end is not None else self.end,
+            show_after=repo.show_after
+            if repo.show_after is not None
+            else self.show_after,
         )
 
 
@@ -136,6 +142,7 @@ def parse_config_file(path: Path) -> MultiRepoConfig:
             min_samples=int(raw_defaults.get("min_samples", 30)),
             start=_optional_str(raw_defaults, "start"),
             end=_optional_str(raw_defaults, "end"),
+            show_after=_optional_str(raw_defaults, "show_after"),
         )
     except (TypeError, ValueError) as e:
         raise ValueError(f"Invalid value in [defaults]: {e}") from e
@@ -166,6 +173,7 @@ def parse_config_file(path: Path) -> MultiRepoConfig:
                 ),
                 start=_optional_str(raw_repo, "start"),
                 end=_optional_str(raw_repo, "end"),
+                show_after=_optional_str(raw_repo, "show_after"),
             )
         except (TypeError, ValueError) as e:
             raise ValueError(
@@ -208,6 +216,7 @@ def parse_config_file(path: Path) -> MultiRepoConfig:
                 ),
                 start=_optional_str(raw_org, "start"),
                 end=_optional_str(raw_org, "end"),
+                show_after=_optional_str(raw_org, "show_after"),
                 exclude_repos=[str(x) for x in exclude],
             )
         except (TypeError, ValueError) as e: