release: Briefprint 0.5.0

CHANGELOG.md

@@ -1,8 +1,6 @@
 # Changelog
 
-## 0.5.0 - Unreleased
-
-Targeted as `0.5.0` because the explicit-use policy is a post-`v0.4.0` routing behavior change.
+## 0.5.0 - 2026-07-06
 
 ### Changed
 
@@ -11,6 +9,7 @@ Targeted as `0.5.0` because the explicit-use policy is a post-`v0.4.0` routing b
 - Added `disable-model-invocation: true` to the installable skill metadata for hosts that support manual-only skill invocation.
 - Documented the boundary between the static agent skill bundle and Briefprint runtime caches under `--cache-dir`.
 - Aligned README cache-dir examples, including benchmark receipts, under `.cache/briefprint`.
+- Bumped runtime and skill metadata to `0.5.0`; cache keys include the skill version, so prior `0.4.x` document-summary and rendered-output cache entries cold-start after upgrade.
 
 ### Validation
 

MANIFEST.in

@@ -5,4 +5,5 @@ recursive-include evals *.json
 recursive-include references *.md
 recursive-include agents *.yaml
 recursive-include docs *.md
+prune docs/superpowers/plans
 recursive-include skills *.md *.yaml

README.ko.md

@@ -143,11 +143,10 @@ python -m document_briefing_cache.cli benchmark \
 | 총 요약 호출 | 4 |
 | 문서 캐시 히트 | 16 |
 | 문서 캐시 미스 | 4 |
-| 품질 경고 행 | 0 |
 
 시나리오별 핵심 결과:
 
-| 시나리오 | 요약 호출 | 캐시 인식 입력 토큰 |
+| 시나리오 | 요약 호출 | 캐시 미스 입력 토큰 |
 |---|---:|---:|
 | 문서 3개 최초 브리핑 | 3 | 226 |
 | 같은 브리핑 반복 | 0 | 0 |
@@ -157,6 +156,8 @@ python -m document_briefing_cache.cli benchmark \
 | 업데이트 1개 추가 | 1 | 83 |
 | 합쳐진 문서 debug 렌더링 | 0 | 0 |
 
+요약 호출과 캐시 미스 입력 토큰이 `0`인 행은 의도된 결과입니다. 반복 브리핑과 재렌더링이 바뀌지 않은 문서를 다시 읽지 않고 캐시된 구조화 상태를 재사용했다는 증거입니다.
+
 숫자는 솔직하게 봐야 합니다. 이 결과는 로컬 harness의 토큰 추정치입니다. 실제 청구 비용을 보려면 provider usage, Codex CLI/OTel, 또는 사용하는 호스트의 telemetry와 함께 비교해야 합니다. 벤치마크의 품질 체크는 actions, decisions, risks, metrics에 대한 가벼운 smoke check이며 의미론적 정확도 점수는 아닙니다.
 
 ## 효과가 큰 경우

README.md

@@ -143,11 +143,10 @@ python -m document_briefing_cache.cli benchmark \
 | Total summarizer calls | 4 |
 | Document cache hits | 16 |
 | Document cache misses | 4 |
-| Quality warning rows | 0 |
 
 Scenario shape:
 
-| Scenario | Summarizer calls | Cache-aware input tokens |
+| Scenario | Summarizer calls | Cache-miss input tokens |
 |---|---:|---:|
 | Cold brief over 3 documents | 3 | 226 |
 | Same brief again | 0 | 0 |
@@ -157,6 +156,8 @@ Scenario shape:
 | Add one update | 1 | 83 |
 | Rerender debug over combined set | 0 | 0 |
 
+Rows with `0` summarizer calls and `0` cache-miss input tokens are intentional: they show that rerenders and repeated briefs reuse cached structured state instead of re-reading unchanged documents.
+
 Honest number warning: these are deterministic local estimates from the benchmark harness, not provider billing telemetry. Use OpenAI/provider usage or host telemetry when exact billing matters. The benchmark also includes lightweight quality smoke checks for obvious actions, decisions, risks, and metrics; it is not a semantic accuracy score.
 
 ## When It Helps

VALIDATION.md

@@ -24,7 +24,7 @@ ruff check .
 - `python scripts\validate_skill.py --run-evals`: `OK: briefprint skill repository validated (24 test files, 9 eval cases, 10 trigger cases, 5 model benchmark cases)`
 - `ruff check .`: `All checks passed!`
 
-Latest fetched `origin/main` CI status was checked separately: GitHub Actions `CI` run `28759630346` completed successfully for `677057a` on 2026-07-06T00:07:29Z. This confirms the base branch status; the commands above remain the source-tree validation for the current checkout.
+Latest fetched `origin/main` CI status was checked separately: GitHub Actions `CI` run `28761659503` completed successfully for `344f05d` on 2026-07-06T01:14:27Z. This confirms the base branch status; the commands above remain the source-tree validation for the current checkout.
 
 `tests/test_distribution_smoke.py` remains opt-in and skips unless `DBC_RUN_INSTALLED_SMOKE=1` is set. CI or release validation should still build wheel and sdist artifacts, install each into a fresh virtual environment, and run the installed smoke test outside the repository root.
 

agents/openai.yaml

@@ -1,4 +1,4 @@
-version: "0.4.0"
+version: "0.5.0"
 
 interface:
   display_name: "Briefprint"

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "document-briefing-cache"
-version = "0.4.0"
+version = "0.5.0"
 description = "Reusable document briefing skill with document-level caching and template rendering."
 requires-python = ">=3.10"
 dependencies = [

scripts/validate_skill.py

@@ -272,7 +272,7 @@ def validate_openai_yaml(path: Path) -> list[str]:
     text = path.read_text(encoding="utf-8")
     errors = []
     required_fragments = [
-        'version: "0.4.0"',
+        'version: "0.5.0"',
         "interface:",
         'display_name: "Briefprint"',
         'short_description: "Rerender cached document briefings without re-summarizing unchanged supplied documents."',

skills/briefprint/agents/openai.yaml

@@ -1,4 +1,4 @@
-version: "0.4.0"
+version: "0.5.0"
 
 interface:
   display_name: "Briefprint"

src/document_briefing_cache/__init__.py

@@ -11,4 +11,4 @@
     "BriefingPipeline",
 ]
 
-__version__ = "0.4.0"
+__version__ = "0.5.0"

src/document_briefing_cache/pipeline.py

@@ -26,7 +26,7 @@
 from .render import TEMPLATE_VERSION, render_briefing
 from .summarizers import BaseSummarizer, RuleBasedExtractiveSummarizer
 
-SKILL_VERSION = "0.4.0"
+SKILL_VERSION = "0.5.0"
 
 
 class BriefingPipeline:

tests/test_skill_metadata.py

@@ -11,14 +11,14 @@
 )
 
 
-def test_versions_are_synchronized_to_0_4_0():
+def test_versions_are_synchronized_to_0_5_0():
     pyproject = (ROOT / "pyproject.toml").read_text(encoding="utf-8")
     openai_yaml = (ROOT / "agents" / "openai.yaml").read_text(encoding="utf-8")
 
-    assert 'version = "0.4.0"' in pyproject
-    assert __version__ == "0.4.0"
-    assert SKILL_VERSION == "0.4.0"
-    assert 'version: "0.4.0"' in openai_yaml
+    assert 'version = "0.5.0"' in pyproject
+    assert __version__ == "0.5.0"
+    assert SKILL_VERSION == "0.5.0"
+    assert 'version: "0.5.0"' in openai_yaml
 
 
 def test_openai_yaml_uses_interface_metadata():