Merge pull request #299 from ggozad/feat/skill-meta

Expose skill state metadata as module-level API

CHANGELOG.md

@@ -1,6 +1,14 @@
 # Changelog
 ## [Unreleased]
 
+### Added
+
+- **Module-level skill introspection API**: `STATE_TYPE`, `STATE_NAMESPACE`, `skill_metadata()`, `instructions()`, and `state_metadata()` on `haiku.rag.skills.rag` and `haiku.rag.skills.rlm` — allows introspecting skill configuration without calling `create_skill()`
+
+### Changed
+
+- **`haiku.skills` dependency**: Bumped to `>=0.7.0` for `StateMetadata` dataclass
+
 ## [0.32.3] - 2026-03-03
 
 ### Changed

haiku_rag_slim/haiku/rag/skills/rag.py

@@ -1,4 +1,5 @@
 import os
+from functools import cache
 from pathlib import Path
 from typing import Any
 
@@ -9,7 +10,7 @@ from haiku.rag.agents.research.models import Citation
 from haiku.rag.store.models.chunk import SearchResult
 from haiku.rag.tools.document import DocumentInfo
 from haiku.rag.tools.qa import QAHistoryEntry
-from haiku.skills.models import Skill, SkillSource
+from haiku.skills.models import Skill, SkillMetadata, SkillSource, StateMetadata
 from haiku.skills.parser import parse_skill_md
 from haiku.skills.state import SkillRunDeps
 
@@ -37,6 +38,32 @@ class RAGState(BaseModel):
     reports: list[ResearchEntry] = Field(default_factory=list)
 
 
+STATE_TYPE = RAGState
+STATE_NAMESPACE = "rag"
+
+_skill_path = Path(__file__).parent / "rag"
+
+
+@cache
+def skill_metadata() -> SkillMetadata:
+    metadata, _ = parse_skill_md(_skill_path / "SKILL.md")
+    return metadata
+
+
+@cache
+def instructions() -> str | None:
+    _, instr = parse_skill_md(_skill_path / "SKILL.md")
+    return instr
+
+
+def state_metadata() -> StateMetadata:
+    return StateMetadata(
+        namespace=STATE_NAMESPACE,
+        type=STATE_TYPE,
+        schema=STATE_TYPE.model_json_schema(),
+    )
+
+
 def create_skill(
     db_path: Path | None = None,
     config: Any = None,
@@ -62,9 +89,6 @@ def create_skill(
         else:
             db_path = config.storage.data_dir / "haiku.rag.lancedb"
 
-    path = Path(__file__).parent / "rag"
-    metadata, instructions = parse_skill_md(path / "SKILL.md")
-
     async def _find_relevant_prior_qa(
         state: RAGState, query: str
     ) -> list[QAHistoryEntry]:
@@ -323,10 +347,10 @@ def create_skill(
         return "\n".join(parts)
 
     return Skill(
-        metadata=metadata,
+        metadata=skill_metadata(),
         source=SkillSource.ENTRYPOINT,
-        path=path,
-        instructions=instructions,
+        path=_skill_path,
+        instructions=instructions(),
         tools=[
             search,
             list_documents,
@@ -334,6 +358,6 @@ def create_skill(
             ask,
             research,
         ],
-        state_type=RAGState,
-        state_namespace="rag",
+        state_type=STATE_TYPE,
+        state_namespace=STATE_NAMESPACE,
     )

haiku_rag_slim/haiku/rag/skills/rlm.py

@@ -1,11 +1,12 @@
 import os
+from functools import cache
 from pathlib import Path
 from typing import Any
 
 from pydantic import BaseModel
 from pydantic_ai import RunContext
 
-from haiku.skills.models import Skill, SkillSource
+from haiku.skills.models import Skill, SkillMetadata, SkillSource, StateMetadata
 from haiku.skills.parser import parse_skill_md
 from haiku.skills.state import SkillRunDeps
 
@@ -20,6 +21,32 @@ class RLMState(BaseModel):
     analyses: list[AnalysisEntry] = []
 
 
+STATE_TYPE = RLMState
+STATE_NAMESPACE = "rlm"
+
+_skill_path = Path(__file__).parent / "rag-rlm"
+
+
+@cache
+def skill_metadata() -> SkillMetadata:
+    metadata, _ = parse_skill_md(_skill_path / "SKILL.md")
+    return metadata
+
+
+@cache
+def instructions() -> str | None:
+    _, instr = parse_skill_md(_skill_path / "SKILL.md")
+    return instr
+
+
+def state_metadata() -> StateMetadata:
+    return StateMetadata(
+        namespace=STATE_NAMESPACE,
+        type=STATE_TYPE,
+        schema=STATE_TYPE.model_json_schema(),
+    )
+
+
 def create_skill(
     db_path: Path | None = None,
     config: Any = None,
@@ -45,9 +72,6 @@ def create_skill(
         else:
             db_path = config.storage.data_dir / "haiku.rag.lancedb"
 
-    path = Path(__file__).parent / "rag-rlm"
-    metadata, instructions = parse_skill_md(path / "SKILL.md")
-
     async def analyze(
         ctx: RunContext[SkillRunDeps],
         question: str,
@@ -85,11 +109,11 @@ def create_skill(
         return output
 
     return Skill(
-        metadata=metadata,
+        metadata=skill_metadata(),
         source=SkillSource.ENTRYPOINT,
-        path=path,
-        instructions=instructions,
+        path=_skill_path,
+        instructions=instructions(),
         tools=[analyze],
-        state_type=RLMState,
-        state_namespace="rlm",
+        state_type=STATE_TYPE,
+        state_namespace=STATE_NAMESPACE,
     )

haiku_rag_slim/pyproject.toml

@@ -24,7 +24,7 @@ classifiers = [
 dependencies = [
     "cachetools>=5.5.0",
     "docling-core==2.65.1",
-    "haiku.skills>=0.6.0",
+    "haiku.skills>=0.7.0",
     "httpx>=0.28.1",
     "jsonpatch>=1.33",
     "lancedb==0.29.2",

tests/skills/test_rag.py

@@ -2,13 +2,56 @@ from unittest.mock import AsyncMock
 
 from haiku.rag.agents.research.models import Citation, ResearchReport
 from haiku.rag.client import HaikuRAG
+from haiku.rag.skills.rag import (
+    STATE_NAMESPACE,
+    STATE_TYPE,
+    RAGState,
+    instructions,
+    skill_metadata,
+    state_metadata,
+)
 from haiku.rag.store.models.chunk import SearchResult
 from haiku.rag.tools.document import DocumentInfo
 from haiku.rag.tools.qa import QAHistoryEntry
+from haiku.skills.models import SkillMetadata, StateMetadata
 
 from .conftest import _get_tool, _make_ctx
 
 
+class TestRAGModuleAPI:
+    def test_state_type_is_rag_state(self):
+        assert STATE_TYPE is RAGState
+
+    def test_state_namespace(self):
+        assert STATE_NAMESPACE == "rag"
+
+    def test_state_metadata_returns_state_metadata(self):
+        result = state_metadata()
+        assert isinstance(result, StateMetadata)
+        assert result.namespace == "rag"
+        assert result.type is RAGState
+        assert result.schema == RAGState.model_json_schema()
+
+    def test_skill_metadata_returns_skill_metadata(self):
+        result = skill_metadata()
+        assert isinstance(result, SkillMetadata)
+        assert result.name == "rag"
+
+    def test_instructions_returns_string(self):
+        result = instructions()
+        assert isinstance(result, str)
+        assert len(result) > 0
+
+    def test_constants_match_create_skill(self, temp_db_path):
+        from haiku.rag.skills.rag import create_skill
+
+        skill = create_skill(db_path=temp_db_path)
+        assert skill.state_type is STATE_TYPE
+        assert skill.state_namespace == STATE_NAMESPACE
+        assert skill.metadata == skill_metadata()
+        assert skill.instructions == instructions()
+
+
 class TestRAGSkillCreation:
     def test_create_skill_returns_valid_skill(self, temp_db_path):
         from haiku.rag.skills.rag import create_skill

tests/skills/test_rlm.py

@@ -2,10 +2,53 @@ from unittest.mock import AsyncMock
 
 from haiku.rag.agents.rlm.models import RLMResult
 from haiku.rag.client import HaikuRAG
+from haiku.rag.skills.rlm import (
+    STATE_NAMESPACE,
+    STATE_TYPE,
+    RLMState,
+    instructions,
+    skill_metadata,
+    state_metadata,
+)
+from haiku.skills.models import SkillMetadata, StateMetadata
 
 from .conftest import _get_tool, _make_ctx
 
 
+class TestRLMModuleAPI:
+    def test_state_type_is_rlm_state(self):
+        assert STATE_TYPE is RLMState
+
+    def test_state_namespace(self):
+        assert STATE_NAMESPACE == "rlm"
+
+    def test_state_metadata_returns_state_metadata(self):
+        result = state_metadata()
+        assert isinstance(result, StateMetadata)
+        assert result.namespace == "rlm"
+        assert result.type is RLMState
+        assert result.schema == RLMState.model_json_schema()
+
+    def test_skill_metadata_returns_skill_metadata(self):
+        result = skill_metadata()
+        assert isinstance(result, SkillMetadata)
+        assert result.name == "rag-rlm"
+
+    def test_instructions_returns_string(self):
+        result = instructions()
+        assert isinstance(result, str)
+        assert len(result) > 0
+
+    def test_constants_match_create_skill(self, temp_db_path):
+        from haiku.rag.skills.rlm import create_skill
+
+        skill = create_skill(db_path=temp_db_path)
+        assert skill.state_type is STATE_TYPE
+        assert skill.state_namespace == STATE_NAMESPACE
+        assert skill.metadata == skill_metadata()
+        assert skill.instructions == instructions()
+
+
 class TestRLMSkillCreation:
     def test_create_skill_returns_valid_skill(self, temp_db_path):
         from haiku.rag.skills.rlm import create_skill

uv.lock

@@ -1490,7 +1490,7 @@ requires-dist = [
     { name = "cohere", marker = "extra == 'cohere'", specifier = ">=5.20.1" },
     { name = "docling", marker = "extra == 'docling'", specifier = "==2.73.1" },
     { name = "docling-core", specifier = "==2.65.1" },
-    { name = "haiku-skills", specifier = ">=0.6.0" },
+    { name = "haiku-skills", specifier = ">=0.7.0" },
     { name = "httpx", specifier = ">=0.28.1" },
     { name = "jsonpatch", specifier = ">=1.33" },
     { name = "lancedb", specifier = "==0.29.2" },
@@ -1522,7 +1522,7 @@ provides-extras = ["docling", "voyageai", "mxbai", "cohere", "zeroentropy", "jin
 
 [[package]]
 name = "haiku-skills"
-version = "0.6.0"
+version = "0.7.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pydantic" },
@@ -1530,9 +1530,9 @@ dependencies = [
     { name = "pyyaml" },
     { name = "skills-ref" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/10/0f/97d95a6814cec171d97ca7eec6365468f6cbdbfc9fbb2d6346494035aa8a/haiku_skills-0.6.0.tar.gz", hash = "sha256:8352c9157260742b475315f92191fe0c9149aa0faa8c9d20af8c201f1bc71e87", size = 132772, upload-time = "2026-03-03T08:55:59.909Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8c/a6/4fbdfe95e6ff6a574088b9f3ed7543a6e01b4b6c1740440ce07c1cf4893a/haiku_skills-0.7.0.tar.gz", hash = "sha256:ec5c5176f8feab09cc6aa4cda9a64e2446074f3ed111cb7e2587f7b524711364", size = 133420, upload-time = "2026-03-04T09:25:52.729Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/4c/c4/f9f8892da06bdc2defb12191e765486dd8dd87155c63dc6ca045f75c8211/haiku_skills-0.6.0-py3-none-any.whl", hash = "sha256:adfbe2206eb238abb0dfe376ffede669860ffd45a2f4647090dfbd787842cf7f", size = 24631, upload-time = "2026-03-03T08:55:58.901Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/38/5048252c0b68bc77dd0fcf6b5c4e29074d90b3f6888eb282a0fcae967227/haiku_skills-0.7.0-py3-none-any.whl", hash = "sha256:7b7791890cc230fd53c5a9a7d93129e7901fecc1d64153d0386a83603e5b7b25", size = 25022, upload-time = "2026-03-04T09:25:51.872Z" },
 ]
 
 [[package]]