kernel-lore-mcp 0.3.1__tar.gz → 0.3.2__tar.gz

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/CHANGELOG.md

@@ -10,6 +10,41 @@ release tags move them into a dated section. Release process in
 
 ## [Unreleased]
 
+## [0.3.2] — 2026-04-22
+
+### Added
+
+- Added `lore_fix_status`, a bug-centric fix-correlation tool that
+  follows `Reported-by:`, `Link:`, and `Closes:` evidence across
+  separate lore threads and upgrades to an authoritative `merged`
+  verdict when the git sidecar can confirm it.
+- Added a normalized trailer-reference side index in `over.db` for
+  `reported_by`, `fixes`, `link`, and `closes`, plus backfill support
+  and reader-side exact lookups for syzbot hashes, lore message-ids,
+  SHA prefixes, lowercased URLs, and normalized trailer payloads.
+- Added human-friendly `since` / `until` parsing for MCP tools and the
+  Rust query router: raw nanoseconds remain supported, but callers can
+  now also pass ISO dates, RFC3339 timestamps, and relative windows
+  such as `90d`.
+
+### Changed
+
+- `lore_activity`, `lore_author_profile`, `lore_file_timeline`,
+  `lore_path_mentions`, `lore_eq`, `lore_in_list`, `lore_count`,
+  `lore_substr_subject`, `lore_substr_trailers`, and `lore_regex` now
+  accept both legacy `*_unix_ns` bounds and new human-readable
+  `since` / `until` fields, with shared validation.
+- `lore_search` query parsing now understands `since:` and `until:`
+  bounds in human formats instead of requiring raw integer nanoseconds
+  only.
+
+### Fixed
+
+- Upper-bound date filtering is now carried consistently through the
+  Rust reader, over.db index helpers, PyO3 boundary, and MCP tool
+  layer instead of being post-filtered inconsistently or silently
+  ignored on some paths.
+
 ## [0.3.1] — 2026-04-22
 
 ### Added

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/Cargo.lock

@@ -2638,7 +2638,7 @@ dependencies = [
 
 [[package]]
 name = "kernel_lore_mcp"
-version = "0.3.1"
+version = "0.3.2"
 dependencies = [
  "aho-corasick",
  "anyhow",

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kernel_lore_mcp"
-version = "0.3.1"
+version = "0.3.2"
 edition = "2024"
 rust-version = "1.85"
 authors = ["Michael Bommarito <michael@bommaritollc.com>"]

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kernel-lore-mcp
-Version: 0.3.1
+Version: 0.3.2
 Classifier: Development Status :: 2 - Pre-Alpha
 Classifier: Intended Audience :: Developers
 Classifier: License :: OSI Approved :: MIT License
@@ -95,7 +95,7 @@ git clone --depth 1 https://github.com/mjbommar/kernel-lore-mcp.git
 cd kernel-lore-mcp && ./scripts/agentic_smoke.sh local
 # PASS: 7/7 tools, 5/5 resource templates, 5/5 prompts (the
 # `REQUIRED_*` subset from src/kernel_lore_mcp/_surface_manifest.py;
-# the live server registers 24 tools in total).
+# the live server registers 25 tools in total).
 ```
 
 Then pick your agent and copy its snippet from
@@ -146,13 +146,14 @@ Want production-grade systemd deployment (single `klmcp-sync.timer`
 replacing the pre-v0.2.0 grokmirror + ingest pair)?
 [`docs/ops/runbook.md`](./docs/ops/runbook.md) §1 onwards.
 
-## Status — v0.3.1 (2026-04-22)
+## Status — v0.3.2 (2026-04-22)
 
-Current release: `v0.3.1`, the first patch line after the hosted
-readiness release. The focus is same-box sync visibility and safer
-inline BM25 behavior: live sync state in `/status`, explicit
-post-ingest stage logs, conservative BM25 writer defaults, and
-load-aware retry hints when writer-heavy work is active.
+Current release: `v0.3.2`, the follow-on patch after the hosted
+readiness and same-box sync hardening line. The focus is better bug
+workflow ergonomics and safer query scoping: a first-class
+`lore_fix_status` tool, indexed trailer-reference correlation for
+syzbot / lore / `Fixes:` joins, and human-readable `since` / `until`
+bounds across both tools and `lore_search`.
 
 Shipped:
 
@@ -164,7 +165,7 @@ Shipped:
   manifest fetch, gix smart-HTTP clone-or-fetch (rayon-fanned
   across shards), ingest, tid rebuild, generation bump — all
   under one writer lock so there's no trigger/debounce race.
-- Full MCP surface: **24 tools** (search, primitives, sampling-
+- Full MCP surface: **25 tools** (search, primitives, sampling-
   backed summarize/classify/explain, authoritative `merged` /
   `picked_up` verdicts via git-sidecar, `lore_corpus_stats` for
   coverage transparency, `lore_author_footprint` for address-

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/README.md

@@ -60,7 +60,7 @@ git clone --depth 1 https://github.com/mjbommar/kernel-lore-mcp.git
 cd kernel-lore-mcp && ./scripts/agentic_smoke.sh local
 # PASS: 7/7 tools, 5/5 resource templates, 5/5 prompts (the
 # `REQUIRED_*` subset from src/kernel_lore_mcp/_surface_manifest.py;
-# the live server registers 24 tools in total).
+# the live server registers 25 tools in total).
 ```
 
 Then pick your agent and copy its snippet from
@@ -111,13 +111,14 @@ Want production-grade systemd deployment (single `klmcp-sync.timer`
 replacing the pre-v0.2.0 grokmirror + ingest pair)?
 [`docs/ops/runbook.md`](./docs/ops/runbook.md) §1 onwards.
 
-## Status — v0.3.1 (2026-04-22)
+## Status — v0.3.2 (2026-04-22)
 
-Current release: `v0.3.1`, the first patch line after the hosted
-readiness release. The focus is same-box sync visibility and safer
-inline BM25 behavior: live sync state in `/status`, explicit
-post-ingest stage logs, conservative BM25 writer defaults, and
-load-aware retry hints when writer-heavy work is active.
+Current release: `v0.3.2`, the follow-on patch after the hosted
+readiness and same-box sync hardening line. The focus is better bug
+workflow ergonomics and safer query scoping: a first-class
+`lore_fix_status` tool, indexed trailer-reference correlation for
+syzbot / lore / `Fixes:` joins, and human-readable `since` / `until`
+bounds across both tools and `lore_search`.
 
 Shipped:
 
@@ -129,7 +130,7 @@ Shipped:
   manifest fetch, gix smart-HTTP clone-or-fetch (rayon-fanned
   across shards), ingest, tid rebuild, generation bump — all
   under one writer lock so there's no trigger/debounce race.
-- Full MCP surface: **24 tools** (search, primitives, sampling-
+- Full MCP surface: **25 tools** (search, primitives, sampling-
   backed summarize/classify/explain, authoritative `merged` /
   `picked_up` verdicts via git-sidecar, `lore_corpus_stats` for
   coverage transparency, `lore_author_footprint` for address-

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/TODO.md

@@ -13,8 +13,22 @@ Status markers:
 ## Current release track (2026-04-22)
 
 `v0.3.0` shipped on 2026-04-21. The active patch line is
-`v0.3.1`: sync-under-load hardening, operator visibility, and release
-cleanup after the first same-box hosted soak on `server6`.
+`v0.3.2`: bug-workflow correlation and human time-bound ergonomics
+after the same-box hosted soak and sync-under-load hardening on
+`server6`.
+
+### 0.3.2 target — bug correlation + human time bounds
+
+- [x] Add a first-class `lore_fix_status` tool for bug-centric
+  correlation across separate lore threads, with authoritative
+  `merged` only when the git sidecar can confirm it.
+- [x] Materialize a normalized trailer-reference index for
+  `reported_by`, `fixes`, `link`, and `closes`, and expose exact
+  reader lookups over syzbot hashes, lore mids, SHA prefixes, and
+  normalized trailer values.
+- [x] Accept human-friendly `since` / `until` values across the
+  low-level tools and query router, while keeping raw
+  `*_unix_ns` compatibility for existing clients.
 
 ### 0.3.1 target — sync-under-load visibility + safer inline BM25
 

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "kernel-lore-mcp"
-version = "0.3.1"
+version = "0.3.2"
 description = "MCP server exposing fast, structured search over lore.kernel.org to LLM developer tools."
 readme = "README.md"
 license = { text = "MIT" }

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/ingest.rs

@@ -30,7 +30,7 @@ use crate::over::{DddPayload, OverDb, OverRow};
 use crate::parse::{self, ParsedMessage};
 use crate::state::State;
 use crate::store::{Store, StoreOffset};
-use crate::trigram::{segment_dir as trigram_segment_dir, SegmentBuilder as TrigramBuilder};
+use crate::trigram::{SegmentBuilder as TrigramBuilder, segment_dir as trigram_segment_dir};
 
 /// Ingest one public-inbox shard end-to-end.
 ///

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/kernel_lore_mcp/__init__.py

@@ -15,7 +15,7 @@ from typing import Any
 
 __all__ = ["__version__", "native_version"]
 
-__version__ = "0.3.1"
+__version__ = "0.3.2"
 
 
 def __getattr__(name: str) -> Any:

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/kernel_lore_mcp/_core.pyi

@@ -33,6 +33,7 @@ def rebuild_tid(data_dir: str | PathLike[str]) -> TidRebuildResult: ...
 def rebuild_bm25(data_dir: str | PathLike[str]) -> int: ...
 def backfill_subject_normalized(data_dir: str | PathLike[str]) -> int: ...
 def backfill_trailer_emails(data_dir: str | PathLike[str]) -> int: ...
+def backfill_trailer_refs(data_dir: str | PathLike[str]) -> int: ...
 def backfill_touched_files(data_dir: str | PathLike[str]) -> int: ...
 def backfill_side_table_dates(data_dir: str | PathLike[str]) -> int: ...
 def rebuild_path_vocab(data_dir: str | PathLike[str]) -> int: ...
@@ -109,6 +110,7 @@ class Reader:
         file: str | None = ...,
         function: str | None = ...,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         list: str | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...
@@ -144,6 +146,7 @@ class Reader:
         field: str,
         value: str,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         list: str | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...
@@ -152,6 +155,7 @@ class Reader:
         field: str,
         values: list[str],
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         list: str | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...
@@ -160,6 +164,7 @@ class Reader:
         field: str,
         value: str,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         list: str | None = ...,
     ) -> dict[str, Any]: ...
     def author_profile(
@@ -167,6 +172,7 @@ class Reader:
         addr: str,
         list: str | None = ...,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         limit: int = ...,
         include_mentions: bool = ...,
         mention_limit: int = ...,
@@ -182,6 +188,7 @@ class Reader:
         needle: str,
         list: str | None = ...,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...
     def substr_trailers(
@@ -190,6 +197,17 @@ class Reader:
         value_substring: str,
         list: str | None = ...,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
+        limit: int = ...,
+    ) -> list[dict[str, Any]]: ...
+    def trailer_ref_lookup(
+        self,
+        name: str,
+        ref_kind: str,
+        ref_value: str,
+        list: str | None = ...,
+        since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...
     def regex(
@@ -199,6 +217,7 @@ class Reader:
         anchor_required: bool = ...,
         list: str | None = ...,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...
     def diff(
@@ -234,5 +253,6 @@ class Reader:
         match_mode: str = ...,
         list: str | None = ...,
         since_unix_ns: int | None = ...,
+        until_unix_ns: int | None = ...,
         limit: int = ...,
     ) -> list[dict[str, Any]]: ...

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/kernel_lore_mcp/server.py

@@ -70,6 +70,7 @@ def build_server(settings: Settings | None = None) -> FastMCP:
     from kernel_lore_mcp.tools.expand_citation import lore_expand_citation
     from kernel_lore_mcp.tools.explain_patch import lore_explain_patch
     from kernel_lore_mcp.tools.file_timeline import lore_file_timeline
+    from kernel_lore_mcp.tools.fix_status import lore_fix_status
     from kernel_lore_mcp.tools.maintainer_profile import lore_maintainer_profile
     from kernel_lore_mcp.tools.message import lore_message
     from kernel_lore_mcp.tools.nearest import lore_nearest, lore_similar
@@ -166,6 +167,10 @@ def build_server(settings: Settings | None = None) -> FastMCP:
         lore_thread_state,
         "Classify a thread (rfc/superseded/nacked/...)",
     )
+    _reg(
+        lore_fix_status,
+        "Correlate a bug report to candidate fixes across threads",
+    )
     _reg(
         lore_subsystem_churn,
         "Hot files in a list/subsystem (top-N + histogram)",

kernel_lore_mcp-0.3.2/src/kernel_lore_mcp/time_bounds.py

@@ -0,0 +1,118 @@
+"""Shared parsing for human-friendly time bounds on MCP tools."""
+
+from __future__ import annotations
+
+import re
+from datetime import UTC, date, datetime, timedelta
+
+from kernel_lore_mcp.errors import invalid_argument
+
+TIME_BOUND_DESCRIPTION = (
+    "ISO date/time, raw nanoseconds since epoch, or a relative window like `90d` / `6mo`."
+)
+
+_RELATIVE_RE = re.compile(r"^(?P<count>\d+)(?P<unit>mo|y|w|d|h|s)$", re.IGNORECASE)
+
+
+def parse_time_bound(
+    *,
+    name: str,
+    value: str,
+    now: datetime | None = None,
+) -> int:
+    raw = value.strip()
+    if not raw:
+        raise invalid_argument(
+            name=name,
+            reason="time bound must be non-empty",
+            value=value,
+            example="2026-01-15 or 90d",
+        )
+    if raw.isdigit():
+        return int(raw)
+
+    match = _RELATIVE_RE.fullmatch(raw)
+    if match:
+        count = int(match.group("count"))
+        unit = match.group("unit").lower()
+        now_dt = now or datetime.now(tz=UTC)
+        if unit == "s":
+            delta = timedelta(seconds=count)
+        elif unit == "h":
+            delta = timedelta(hours=count)
+        elif unit == "d":
+            delta = timedelta(days=count)
+        elif unit == "w":
+            delta = timedelta(weeks=count)
+        elif unit == "mo":
+            delta = timedelta(days=30 * count)
+        else:
+            delta = timedelta(days=365 * count)
+        return int((now_dt - delta).timestamp() * 1_000_000_000)
+
+    try:
+        if len(raw) == 10 and raw.count("-") == 2:
+            dt = datetime.combine(date.fromisoformat(raw), datetime.min.time(), tzinfo=UTC)
+        else:
+            dt = datetime.fromisoformat(raw.replace("Z", "+00:00"))
+            if dt.tzinfo is None:
+                dt = dt.replace(tzinfo=UTC)
+            else:
+                dt = dt.astimezone(UTC)
+    except ValueError as exc:
+        raise invalid_argument(
+            name=name,
+            reason=(
+                "time bound must be ISO-8601 (`2026-01-15`, `2026-01-15T00:00:00Z`), "
+                "raw nanoseconds, or a relative window like `90d` / `6mo`"
+            ),
+            value=value,
+            example="2026-01-15 or 90d",
+        ) from exc
+    return int(dt.timestamp() * 1_000_000_000)
+
+
+def resolve_time_bounds(
+    *,
+    since: str | None = None,
+    since_unix_ns: int | None = None,
+    until: str | None = None,
+    until_unix_ns: int | None = None,
+) -> tuple[int | None, int | None]:
+    if since is not None and since_unix_ns is not None:
+        raise invalid_argument(
+            name="since",
+            reason="pass only one of `since` or `since_unix_ns`",
+            value={"since": since, "since_unix_ns": since_unix_ns},
+            example='{"since": "2026-01-15"}',
+        )
+    if until is not None and until_unix_ns is not None:
+        raise invalid_argument(
+            name="until",
+            reason="pass only one of `until` or `until_unix_ns`",
+            value={"until": until, "until_unix_ns": until_unix_ns},
+            example='{"until": "2026-04-01"}',
+        )
+
+    resolved_since = since_unix_ns
+    resolved_until = until_unix_ns
+    if since is not None:
+        resolved_since = parse_time_bound(name="since", value=since)
+    if until is not None:
+        resolved_until = parse_time_bound(name="until", value=until)
+
+    if (
+        resolved_since is not None
+        and resolved_until is not None
+        and resolved_since >= resolved_until
+    ):
+        raise invalid_argument(
+            name="window",
+            reason="since must be less than until",
+            value={"since": resolved_since, "until": resolved_until},
+            example='{"since": "2026-01-01", "until": "2026-02-01"}',
+        )
+    return resolved_since, resolved_until
+
+
+__all__ = ["TIME_BOUND_DESCRIPTION", "parse_time_bound", "resolve_time_bounds"]

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/kernel_lore_mcp/tools/activity.py

@@ -15,6 +15,7 @@ from kernel_lore_mcp.cursor import decode_cursor, mint_cursor, query_hash
 from kernel_lore_mcp.freshness import build_freshness
 from kernel_lore_mcp.mapping import row_to_activity_row
 from kernel_lore_mcp.models import ActivityResponse
+from kernel_lore_mcp.time_bounds import TIME_BOUND_DESCRIPTION, resolve_time_bounds
 from kernel_lore_mcp.timeout import run_with_timeout
 
 _CONCISE_ROWS = 20
@@ -29,10 +30,22 @@ async def lore_activity(
         str | None,
         Field(description="Exact identifier, e.g. `smb_check_perm_dacl`."),
     ] = None,
+    since: Annotated[
+        str | None,
+        Field(description=f"Human-friendly lower bound. {TIME_BOUND_DESCRIPTION}"),
+    ] = None,
     since_unix_ns: Annotated[
         int | None,
         Field(description="Lower bound on message date (nanoseconds since epoch UTC)."),
     ] = None,
+    until: Annotated[
+        str | None,
+        Field(description=f"Human-friendly exclusive upper bound. {TIME_BOUND_DESCRIPTION}"),
+    ] = None,
+    until_unix_ns: Annotated[
+        int | None,
+        Field(description="Exclusive upper bound on message date (nanoseconds since epoch UTC)."),
+    ] = None,
     list: Annotated[
         str | None,
         Field(description="Restrict to one mailing list (e.g. `linux-cifs`)."),
@@ -78,6 +91,12 @@ async def lore_activity(
 
     settings = get_settings()
     reader = _core.Reader(settings.data_dir)
+    resolved_since, resolved_until = resolve_time_bounds(
+        since=since,
+        since_unix_ns=since_unix_ns,
+        until=until,
+        until_unix_ns=until_unix_ns,
+    )
 
     # Cursor scope: (file, function, since, list) define the sorted
     # result set. `limit` / `response_format` can change between
@@ -86,7 +105,8 @@ async def lore_activity(
         "lore_activity",
         file or "",
         function or "",
-        since_unix_ns or 0,
+        resolved_since or 0,
+        resolved_until or 0,
         list or "",
     )
     resume = decode_cursor(cursor, expected_q_hash=q_hash, arg_name="cursor")
@@ -98,7 +118,8 @@ async def lore_activity(
         reader.activity,
         file,
         function,
-        since_unix_ns,
+        resolved_since,
+        resolved_until,
         list,
         fetch_budget,
     )

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/kernel_lore_mcp/tools/author_footprint.py

@@ -66,9 +66,7 @@ class FootprintHit(BaseModel):
 
 class AuthorFootprintResponse(BaseModel):
     addr_queried: str
-    total_distinct: int = Field(
-        description="Distinct message_ids across all three sources."
-    )
+    total_distinct: int = Field(description="Distinct message_ids across all three sources.")
     authored_count: int
     trailer_mention_count: int
     body_mention_count: int
@@ -165,7 +163,7 @@ async def lore_author_footprint(
 
     # Source 1: authored rows, via indexed from_addr.
     authored_rows = await run_with_timeout(
-        reader.eq, "from_addr", addr, None, list_filter, limit
+        reader.eq, "from_addr", addr, None, None, list_filter, limit
     )
     authored_mids = {r["message_id"] for r in authored_rows}
 
@@ -182,6 +180,7 @@ async def lore_author_footprint(
                 addr,
                 list_filter,
                 None,
+                None,
                 0,  # we don't want the authored slice here
                 True,  # include_mentions
                 min(limit, 500),
@@ -209,11 +208,7 @@ async def lore_author_footprint(
         query = " ".join(tokens)
         try:
             scored = await run_with_timeout(reader.prose_search, query, limit)
-            body_rows = [
-                (r, r.get("_score", 0.0))
-                for r in scored
-                if r.get("message_id")
-            ]
+            body_rows = [(r, r.get("_score", 0.0)) for r in scored if r.get("message_id")]
         except Exception:  # noqa: BLE001
             # BM25 can still error on pathological inputs; skipping
             # this source is better than 500ing the whole tool.
@@ -241,11 +236,7 @@ async def lore_author_footprint(
     # Optional list_filter post-filter on body_mention rows
     # (BM25 doesn't honor it intrinsically).
     if list_filter is not None:
-        per_mid = {
-            mid: row
-            for mid, row in per_mid.items()
-            if row.get("list") == list_filter
-        }
+        per_mid = {mid: row for mid, row in per_mid.items() if row.get("list") == list_filter}
         roles = {mid: roles[mid] for mid in per_mid}
 
     # Sort newest-first; cursor-skip before truncating.
@@ -283,15 +274,9 @@ async def lore_author_footprint(
         for row in page
     ]
 
-    authored_count = sum(
-        1 for h in hits if "authored" in h.roles
-    )
-    trailer_count = sum(
-        1 for h in hits if "trailer_mention" in h.roles
-    )
-    body_count = sum(
-        1 for h in hits if "body_mention" in h.roles
-    )
+    authored_count = sum(1 for h in hits if "authored" in h.roles)
+    trailer_count = sum(1 for h in hits if "trailer_mention" in h.roles)
+    body_count = sum(1 for h in hits if "body_mention" in h.roles)
 
     next_cursor: str | None = None
     if hits and total_available > limit:

{kernel_lore_mcp-0.3.1 → kernel_lore_mcp-0.3.2}/src/kernel_lore_mcp/tools/author_profile.py

@@ -33,6 +33,7 @@ from kernel_lore_mcp.models import (
     ReceivedTrailerStats,
     SubsystemBucket,
 )
+from kernel_lore_mcp.time_bounds import TIME_BOUND_DESCRIPTION, resolve_time_bounds
 from kernel_lore_mcp.timeout import run_with_timeout
 
 
@@ -64,10 +65,22 @@ async def lore_author_profile(
             ),
         ),
     ] = None,
+    since: Annotated[
+        str | None,
+        Field(description=f"Human-friendly lower bound. {TIME_BOUND_DESCRIPTION}"),
+    ] = None,
     since_unix_ns: Annotated[
         int | None,
         Field(description="Lower-bound on message date (ns since epoch)."),
     ] = None,
+    until: Annotated[
+        str | None,
+        Field(description=f"Human-friendly exclusive upper bound. {TIME_BOUND_DESCRIPTION}"),
+    ] = None,
+    until_unix_ns: Annotated[
+        int | None,
+        Field(description="Exclusive upper-bound on message date (ns since epoch)."),
+    ] = None,
     limit: Annotated[
         int,
         Field(
@@ -128,18 +141,27 @@ async def lore_author_profile(
             value=addr,
             example="gregkh@linuxfoundation.org",
         )
-    if include_mentions and list_filter is None and since_unix_ns is None:
+    resolved_since, resolved_until = resolve_time_bounds(
+        since=since,
+        since_unix_ns=since_unix_ns,
+        until=until,
+        until_unix_ns=until_unix_ns,
+    )
+    if include_mentions and list_filter is None and resolved_since is None:
         raise invalid_argument(
             name="include_mentions",
             reason=(
                 "unbounded mention scan is not allowed on this server; "
                 "pass either `list_filter` (e.g. list_filter='linux-cifs') "
-                "or `since_unix_ns` (a date lower bound) to narrow"
-            ),
-            value={"include_mentions": True, "list_filter": None, "since_unix_ns": None},
-            example=(
-                '{"include_mentions": true, "list_filter": "linux-cifs"}'
+                "or `since` / `since_unix_ns` (a date lower bound) to narrow"
             ),
+            value={
+                "include_mentions": True,
+                "list_filter": None,
+                "since": since,
+                "since_unix_ns": since_unix_ns,
+            },
+            example=('{"include_mentions": true, "list_filter": "linux-cifs"}'),
         )
 
     from kernel_lore_mcp import _core
@@ -150,7 +172,8 @@ async def lore_author_profile(
         reader.author_profile,
         addr,
         list_filter,
-        since_unix_ns,
+        resolved_since,
+        resolved_until,
         limit,
         include_mentions,
         mention_limit,