scholarx 2.3.4__tar.gz → 2.3.6__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (68) hide show
  1. {scholarx-2.3.4/scholarx.egg-info → scholarx-2.3.6}/PKG-INFO +4 -5
  2. {scholarx-2.3.4 → scholarx-2.3.6}/README.md +0 -1
  3. {scholarx-2.3.4 → scholarx-2.3.6}/pyproject.toml +7 -7
  4. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/api_client.py +1 -1
  5. scholarx-2.3.6/scholarx/connector_sources.py +101 -0
  6. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/mcp_server.py +2 -1
  7. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/certification.json +4 -7
  8. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/scholarx.ttl +4 -2
  9. scholarx-2.3.6/scholarx/skills/scholarx-development/SKILL.md +84 -0
  10. {scholarx-2.3.4 → scholarx-2.3.6/scholarx.egg-info}/PKG-INFO +4 -5
  11. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx.egg-info/SOURCES.txt +2 -2
  12. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx.egg-info/entry_points.txt +0 -3
  13. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx.egg-info/requires.txt +2 -2
  14. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_cli_extended.py +4 -2
  15. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_kg_ingest.py +1 -1
  16. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_mcp_server.py +2 -6
  17. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_operation_tools.py +1 -27
  18. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_parallel_download.py +2 -6
  19. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_queue.py +7 -7
  20. scholarx-2.3.4/scholarx/connectors/mcp_source_presets.json +0 -14
  21. scholarx-2.3.4/scholarx/connectors/tool_schema_fingerprints.json +0 -8
  22. {scholarx-2.3.4 → scholarx-2.3.6}/LICENSE +0 -0
  23. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/__init__.py +0 -0
  24. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/__main__.py +0 -0
  25. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/cli.py +0 -0
  26. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/connectors/__init__.py +0 -0
  27. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/deduplication.py +0 -0
  28. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/kg_ingest.py +0 -0
  29. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/kg_media.py +0 -0
  30. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/mcp_config.json +0 -0
  31. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/models.py +0 -0
  32. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/__init__.py +0 -0
  33. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/fixtures/records.json +0 -0
  34. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/mappings/source.yaml +0 -0
  35. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/migrations/manifest.json +0 -0
  36. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/ontology/shapes/connector.shacl.ttl +0 -0
  37. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/paper_storage.py +0 -0
  38. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/prompts/__init__.py +0 -0
  39. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/__init__.py +0 -0
  40. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/arxiv.py +0 -0
  41. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/base.py +0 -0
  42. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/biorxiv.py +0 -0
  43. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/osf.py +0 -0
  44. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/pmc.py +0 -0
  45. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/rss.py +0 -0
  46. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/providers/semantic_scholar.py +0 -0
  47. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/queue.py +0 -0
  48. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/__init__.py +0 -0
  49. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-kg-ingestion/SKILL.md +0 -0
  50. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-kg-ingestion/WORKFLOW.md +0 -0
  51. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-operations/SKILL.md +0 -0
  52. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-operations/agents/openai.yaml +0 -0
  53. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-operations/references/catalog.md +0 -0
  54. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-paper-library/SKILL.md +0 -0
  55. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-paper-library/WORKFLOW.md +0 -0
  56. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-paper-search/SKILL.md +0 -0
  57. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx/skills/scholarx-paper-search/WORKFLOW.md +0 -0
  58. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx.egg-info/dependency_links.txt +0 -0
  59. {scholarx-2.3.4 → scholarx-2.3.6}/scholarx.egg-info/top_level.txt +0 -0
  60. {scholarx-2.3.4 → scholarx-2.3.6}/setup.cfg +0 -0
  61. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_action_discovery.py +0 -0
  62. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_cli.py +0 -0
  63. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_default_content_surfaces.py +0 -0
  64. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_init_dynamics.py +0 -0
  65. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_kg_media.py +0 -0
  66. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_models.py +0 -0
  67. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_paper_storage.py +0 -0
  68. {scholarx-2.3.4 → scholarx-2.3.6}/tests/test_providers.py +0 -0
@@ -1,13 +1,13 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: scholarx
3
- Version: 2.3.4
3
+ Version: 2.3.6
4
4
  Summary: Universal Research Paper API — single entry point for arXiv, PMC, bioRxiv, medRxiv, PsyArXiv, OSF, and Semantic Scholar
5
- Author-email: Project Maintainers <maintainers@users.noreply.github.com>
5
+ Author-email: Repository Maintainers <maintainers@knuckles.team>
6
6
  License: MIT
7
7
  Requires-Python: <3.15,>=3.12
8
8
  Description-Content-Type: text/markdown
9
9
  License-File: LICENSE
10
- Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.3.18
10
+ Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.4.1
11
11
  Requires-Dist: httpx>=0.28.1
12
12
  Requires-Dist: python-dotenv>=1.2.4
13
13
  Requires-Dist: Levenshtein>=0.27.5
@@ -15,7 +15,7 @@ Requires-Dist: pypdf>=6.21.0
15
15
  Requires-Dist: rich>=15.0.0
16
16
  Requires-Dist: defusedxml>=0.7.1
17
17
  Provides-Extra: mcp
18
- Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.3.18; extra == "mcp"
18
+ Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.4.1; extra == "mcp"
19
19
  Provides-Extra: all
20
20
  Requires-Dist: scholarx[mcp]; extra == "all"
21
21
  Provides-Extra: test
@@ -48,7 +48,6 @@ Dynamic: license-file
48
48
  ![PyPI - Wheel](https://img.shields.io/pypi/wheel/scholarx)
49
49
  ![PyPI - Implementation](https://img.shields.io/pypi/implementation/scholarx)
50
50
 
51
- *Version: 2.3.4*
52
51
 
53
52
  > **Documentation** — Installation, deployment, usage across the API, CLI, MCP, and
54
53
  > agent interfaces are maintained in the
@@ -20,7 +20,6 @@
20
20
  ![PyPI - Wheel](https://img.shields.io/pypi/wheel/scholarx)
21
21
  ![PyPI - Implementation](https://img.shields.io/pypi/implementation/scholarx)
22
22
 
23
- *Version: 2.3.4*
24
23
 
25
24
  > **Documentation** — Installation, deployment, usage across the API, CLI, MCP, and
26
25
  > agent interfaces are maintained in the
@@ -4,14 +4,14 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "scholarx"
7
- version = "2.3.4"
7
+ version = "2.3.6"
8
8
  description = "Universal Research Paper API — single entry point for arXiv, PMC, bioRxiv, medRxiv, PsyArXiv, OSF, and Semantic Scholar"
9
9
  readme = "README.md"
10
10
  license = { text = "MIT" }
11
11
  requires-python = ">=3.12, <3.15"
12
- authors = [{ name = "Project Maintainers", email = "maintainers@users.noreply.github.com" }]
12
+ authors = [{ name = "Repository Maintainers", email = "maintainers@knuckles.team" }]
13
13
  dependencies = [
14
- "agent-connector-sdk[mcp]>=0.3.18,<1.0.0",
14
+ "agent-connector-sdk[mcp]>=0.4.1,<1.0.0",
15
15
  "httpx>=0.28.1",
16
16
  "python-dotenv>=1.2.4",
17
17
  "Levenshtein>=0.27.5",
@@ -21,7 +21,7 @@ dependencies = [
21
21
  ]
22
22
 
23
23
  [project.optional-dependencies]
24
- mcp = ["agent-connector-sdk[mcp]>=0.3.18,<1.0.0"]
24
+ mcp = ["agent-connector-sdk[mcp]>=0.4.1,<1.0.0"]
25
25
  all = ["scholarx[mcp]"]
26
26
  test = [
27
27
  "pytest-xdist>=3.8.0","pytest>=9.1.1", "pytest-asyncio>=1.4.0", "pytest-cov>=7.1.0", "respx>=0.23.1"]
@@ -36,9 +36,6 @@ scholarx = "scholarx.skills"
36
36
  [project.entry-points."agent_connector_sdk.ontology_providers"]
37
37
  scholarx = "scholarx.ontology"
38
38
 
39
- [project.entry-points."agent_connector_sdk.source_connector_providers"]
40
- scholarx = "scholarx.connectors"
41
-
42
39
  # The one declared connector identity: the fleet name (SDK-CONNECTOR-CONTROL-R071).
43
40
  [project.entry-points."agent_connector_sdk.connector"]
44
41
  scholarx-mcp = "scholarx"
@@ -79,3 +76,6 @@ ignore_names = ["request", "config"]
79
76
 
80
77
  [tool.pipelines_hooks.ste]
81
78
  exempt = ["specs/*", "specs/**/*"]
79
+
80
+ [tool.pipelines.api_integration]
81
+ baseline = 50.0
@@ -80,7 +80,7 @@ def _schedule_download_tasks(normalized: list[str | None], one_fn) -> list[async
80
80
 
81
81
 
82
82
  def _normalize_download_ids(ids: list[str]) -> list[str | None]:
83
- """Normalize each raw id, keeping position; an unparseable id becomes None."""
83
+ """Normalize each raw id, keeping position; an unparsable id becomes None."""
84
84
  normalized: list[str | None] = []
85
85
  for raw_id in ids:
86
86
  try:
@@ -0,0 +1,101 @@
1
+ """Data sources of this connector, declared by tool name (generated by connector-ontology-map migrate).
2
+
3
+ Served live as ``connector://resources``. Regenerate by editing the declarations, not the tools.
4
+ """
5
+
6
+ # ruff: noqa: E501
7
+ from __future__ import annotations
8
+
9
+ from agent_connector_sdk.ingest.live_declaration import source_meta
10
+ from agent_connector_sdk.ingest.sources import IngestSourceDeclaration, IngestSourceSpec
11
+
12
+ CONNECTOR = "scholarx-mcp"
13
+
14
+ SOURCES = (
15
+ source_meta(
16
+ tool="scholarx_search",
17
+ source_id="scholarx-papers",
18
+ record_type="research_paper",
19
+ id_field="id",
20
+ cursor_field="published_date",
21
+ preset={
22
+ "doc_type": "research_paper",
23
+ "params_style": "args",
24
+ },
25
+ records_path="papers",
26
+ server="scholarx-mcp",
27
+ text_field="abstract",
28
+ title_field="title",
29
+ ),
30
+ )
31
+
32
+ SCHEMA_MAPPINGS = {
33
+ "Paper": {
34
+ "fields": {
35
+ "abstract": "xsd:string",
36
+ "citationCount": "xsd:integer",
37
+ "doi": "xsd:string",
38
+ "externalToolId": "xsd:string",
39
+ "pdfUrl": "xsd:string",
40
+ "publishedDate": "xsd:string",
41
+ },
42
+ "ontology_class": "Document",
43
+ },
44
+ "PaperSource": {
45
+ "fields": {
46
+ "abstract": "xsd:string",
47
+ "citationCount": "xsd:integer",
48
+ "doi": "xsd:string",
49
+ "externalToolId": "xsd:string",
50
+ "pdfUrl": "xsd:string",
51
+ "publishedDate": "xsd:string",
52
+ },
53
+ "ontology_class": None,
54
+ },
55
+ "ResearchCategory": {
56
+ "fields": {
57
+ "abstract": "xsd:string",
58
+ "citationCount": "xsd:integer",
59
+ "doi": "xsd:string",
60
+ "externalToolId": "xsd:string",
61
+ "pdfUrl": "xsd:string",
62
+ "publishedDate": "xsd:string",
63
+ },
64
+ "ontology_class": None,
65
+ },
66
+ }
67
+
68
+ RESOURCES = (
69
+ {
70
+ "id_prefix": "paper",
71
+ "label": "Research Paper",
72
+ "name": "Paper",
73
+ },
74
+ {
75
+ "id_prefix": "papersource",
76
+ "label": "Paper Source",
77
+ "name": "PaperSource",
78
+ },
79
+ {
80
+ "id_prefix": "researchcategory",
81
+ "label": "Research Category",
82
+ "name": "ResearchCategory",
83
+ },
84
+ )
85
+
86
+
87
+ def declaration() -> IngestSourceDeclaration:
88
+ """The declaration the server assembles from this module."""
89
+ specs = []
90
+ for item in SOURCES:
91
+ for entry in item["connector_sources"]:
92
+ body = {k: v for k, v in entry.items() if k not in ("tool", "mapping")}
93
+ specs.append(IngestSourceSpec.model_validate({**body, "tools": (entry["tool"],)}))
94
+ return IngestSourceDeclaration.model_validate(
95
+ {
96
+ "connector": CONNECTOR,
97
+ "sources": specs,
98
+ "schema_mappings": SCHEMA_MAPPINGS,
99
+ "resources": RESOURCES,
100
+ }
101
+ )
@@ -9,6 +9,7 @@ import asyncio
9
9
  import logging
10
10
  import sys
11
11
  from dataclasses import dataclass
12
+ from importlib.metadata import version as _distribution_version
12
13
 
13
14
  from agent_connector_sdk.config import load_config, setting
14
15
  from agent_connector_sdk.mcp.action_dispatch import resolve_action
@@ -19,7 +20,7 @@ from pydantic import Field
19
20
 
20
21
  load_config()
21
22
 
22
- __version__ = "2.3.4"
23
+ __version__ = _distribution_version("scholarx")
23
24
 
24
25
  # Wall budget for an inline single-paper download before it is handed to the
25
26
  # background queue. Kept well under the MCP child-call timeout so a slow source
@@ -1,13 +1,10 @@
1
1
  {
2
- "api_version": "arbora.io/v1",
2
+ "api_version": "graphos.io/v1",
3
3
  "artifacts": {
4
- "connector_manifest.yml": "d88fbe9425a4161a1437334f0120df1c8f1aeb701c90fbe1bf9bdaa308a052e9",
5
- "scholarx/connectors/mcp_source_presets.json": "e2a46d1357825817c792e9329a58b4e4af33c3147ed9c66fec337584b098ff86",
6
- "scholarx/connectors/tool_schema_fingerprints.json": "83976eb6a8495978c3df1ea72ce7ead7fcd3d567228343301ab55721e10f3fa8",
7
4
  "scholarx/ontology/fixtures/records.json": "b0388efa7f3896eb816a9dd6ff681177e35e08f8cef1b36b50574b539e7c7500",
8
5
  "scholarx/ontology/mappings/source.yaml": "18a3036f8139108be4ceefab417c04e7232d0f532bad4cbdf392c503664598dc",
9
6
  "scholarx/ontology/migrations/manifest.json": "bfeda48a1409171c681c100c996615d1710e6a56235d52e0e99572d75d1aa892",
10
- "scholarx/ontology/scholarx.ttl": "700dc4a81863ca578865190cf1dae85149265bdc0fad6271100a70f3a6de3098",
7
+ "scholarx/ontology/scholarx.ttl": "7956ac7953d8f2649d22981d74ef98cfad3d5c19ee55540e0e2f632222c9b88b",
11
8
  "scholarx/ontology/shapes/connector.shacl.ttl": "2e223e425a72b24e2227f98f0692baf22c0476ae073febb05e4d5fc7e75be438"
12
9
  },
13
10
  "checks": {
@@ -18,7 +15,7 @@
18
15
  "synthetic_fixture_contract": "passed"
19
16
  },
20
17
  "compatibility": {
21
- "agent_connector_sdk": ">=2.5.0,<3",
18
+ "agent_utilities": ">=2.5.0,<3",
22
19
  "bundle_schema": "2",
23
20
  "epistemic_graph": ">=2.27.0,<3"
24
21
  },
@@ -32,5 +29,5 @@
32
29
  "signer": "UNSIGNED-PREVIEW",
33
30
  "signing_public_key": "",
34
31
  "status": "source-validated",
35
- "validated_at": "2026-08-26T13:42:55Z"
32
+ "validated_at": "2026-10-26T16:42:10Z"
36
33
  }
@@ -104,5 +104,7 @@
104
104
  source repositories, subject categories, authors (shared :Person), abstracts and
105
105
  full-text PDFs (shared :Document / :AssetOccurrence) — federated into the agent-connector-sdk
106
106
  knowledge-graph hub across arXiv, PMC, bioRxiv, medRxiv, PsyArXiv, OSF and
107
- Semantic Scholar.""" ;
108
- owl:imports <http://knuckles.team/kg> .
107
+ Semantic Scholar.""" .
108
+
109
+ # ontology-codemod R3
110
+ <http://knuckles.team/kg#ResearchCategory> rdfs:subClassOf <http://knuckles.team/kg#Category> .
@@ -0,0 +1,84 @@
1
+ ---
2
+ name: scholarx-development
3
+ skill_type: skill
4
+ domain: development
5
+ description: >-
6
+ Develop, test, gate, and release the scholarx connector. Use this skill
7
+ when changing paper provider mappings, MCP tools, storage, packaging,
8
+ documentation, or repository gates.
9
+ license: MIT
10
+ tags: [scholarx, development, connector, test, gates]
11
+ tier: repository
12
+ metadata:
13
+ author: Genius
14
+ version: '0.1.0'
15
+ ---
16
+
17
+ # scholarx development
18
+
19
+ This repository owns the ScholarX paper-provider clients, deduplication and
20
+ storage behavior, MCP tools, packaged connector content, and agent entry
21
+ points. Shared connector contracts and transports belong to
22
+ `agent-connector-sdk`; admission, serving, and durable graph state belong to
23
+ their owning repositories. Read `AGENTS.md` before changing code and keep
24
+ provider credentials, paper paths, and endpoints out of source, logs, and
25
+ skill content.
26
+
27
+ ## Rapid delivery: the 5-minute contract
28
+
29
+ Specs are already designed. Deliver small, reviewable slices and cite the
30
+ owning spec row in the commit trailer.
31
+
32
+ - Find the owning row first with `git grep -n "<term>" -- specs` and extend
33
+ that spec before adding new requirement text.
34
+ - Reuse the existing `ScholarXClient`, provider rate limits, deduplication,
35
+ storage, and SDK tool surface. Do not create a second implementation of a
36
+ connector capability.
37
+ - Keep changes scoped to this connector. A change owned by the SDK, Arbora, or
38
+ the knowledge graph belongs in that repository instead.
39
+ - Every delivered requirement needs a bound test. Generated `status.json`
40
+ files are produced by `pipelines-hook spec-status`; never hand-edit them.
41
+ - Mock external paper-provider calls and keep download tests within the
42
+ existing queue and timeout seams; never use live credentials in a local gate.
43
+ - One logical change per commit, with `Spec: <ID>` and the assisting agent
44
+ trailer adjacent in the final paragraph.
45
+
46
+ ## Setup and gates
47
+
48
+ Use the pinned environment from `uv sync --locked`. Run the repository gate:
49
+
50
+ ```bash
51
+ uvx pre-commit run --all-files
52
+ ```
53
+
54
+ For focused feedback, run the smallest affected test file with the locked
55
+ environment, for example:
56
+
57
+ ```bash
58
+ uv run --all-extras pytest tests/test_operation_tools.py -q
59
+ uv run --all-extras pytest tests/test_action_discovery.py -q
60
+ ```
61
+
62
+ The canonical `.pre-commit-config.yaml` delegates connector-standard, secrets,
63
+ spec, tests-fast, and dependency-readiness checks to `Knuckles-Team/pipelines`
64
+ at `rev: main`. Do not replace those shared hooks with local copies or bypass a
65
+ failing check.
66
+
67
+ ## Connector-specific checks
68
+
69
+ - Keep provider fan-out, rate limiting, deduplication, storage, and MCP tool
70
+ surfaces aligned with the SDK.
71
+ - Use mocked arXiv, PMC, bioRxiv, OSF, and Semantic Scholar responses in tests;
72
+ never call live providers from a local gate.
73
+ - Keep `scholarx/skills/`, ontology, prompts, and connector manifests in
74
+ package data when adding content.
75
+ - Documentation lives under `docs/` and is published by the shared Pages
76
+ workflow; source status records live under `specs/`.
77
+
78
+ ## Delivery discipline
79
+
80
+ Work in a real Git worktree based on `origin/main`. Inspect the complete diff,
81
+ stage only reviewed paths, and verify `git status --short` after every gate.
82
+ Do not use `git stash`, `git add -A`, or `git add .`; do not claim Pages
83
+ publication until a successful exact-source workflow run and the public page
84
+ are both observed.
@@ -1,13 +1,13 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: scholarx
3
- Version: 2.3.4
3
+ Version: 2.3.6
4
4
  Summary: Universal Research Paper API — single entry point for arXiv, PMC, bioRxiv, medRxiv, PsyArXiv, OSF, and Semantic Scholar
5
- Author-email: Project Maintainers <maintainers@users.noreply.github.com>
5
+ Author-email: Repository Maintainers <maintainers@knuckles.team>
6
6
  License: MIT
7
7
  Requires-Python: <3.15,>=3.12
8
8
  Description-Content-Type: text/markdown
9
9
  License-File: LICENSE
10
- Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.3.18
10
+ Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.4.1
11
11
  Requires-Dist: httpx>=0.28.1
12
12
  Requires-Dist: python-dotenv>=1.2.4
13
13
  Requires-Dist: Levenshtein>=0.27.5
@@ -15,7 +15,7 @@ Requires-Dist: pypdf>=6.21.0
15
15
  Requires-Dist: rich>=15.0.0
16
16
  Requires-Dist: defusedxml>=0.7.1
17
17
  Provides-Extra: mcp
18
- Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.3.18; extra == "mcp"
18
+ Requires-Dist: agent-connector-sdk[mcp]<1.0.0,>=0.4.1; extra == "mcp"
19
19
  Provides-Extra: all
20
20
  Requires-Dist: scholarx[mcp]; extra == "all"
21
21
  Provides-Extra: test
@@ -48,7 +48,6 @@ Dynamic: license-file
48
48
  ![PyPI - Wheel](https://img.shields.io/pypi/wheel/scholarx)
49
49
  ![PyPI - Implementation](https://img.shields.io/pypi/implementation/scholarx)
50
50
 
51
- *Version: 2.3.4*
52
51
 
53
52
  > **Documentation** — Installation, deployment, usage across the API, CLI, MCP, and
54
53
  > agent interfaces are maintained in the
@@ -5,6 +5,7 @@ scholarx/__init__.py
5
5
  scholarx/__main__.py
6
6
  scholarx/api_client.py
7
7
  scholarx/cli.py
8
+ scholarx/connector_sources.py
8
9
  scholarx/deduplication.py
9
10
  scholarx/kg_ingest.py
10
11
  scholarx/kg_media.py
@@ -20,8 +21,6 @@ scholarx.egg-info/entry_points.txt
20
21
  scholarx.egg-info/requires.txt
21
22
  scholarx.egg-info/top_level.txt
22
23
  scholarx/connectors/__init__.py
23
- scholarx/connectors/mcp_source_presets.json
24
- scholarx/connectors/tool_schema_fingerprints.json
25
24
  scholarx/ontology/__init__.py
26
25
  scholarx/ontology/certification.json
27
26
  scholarx/ontology/scholarx.ttl
@@ -39,6 +38,7 @@ scholarx/providers/pmc.py
39
38
  scholarx/providers/rss.py
40
39
  scholarx/providers/semantic_scholar.py
41
40
  scholarx/skills/__init__.py
41
+ scholarx/skills/scholarx-development/SKILL.md
42
42
  scholarx/skills/scholarx-kg-ingestion/SKILL.md
43
43
  scholarx/skills/scholarx-kg-ingestion/WORKFLOW.md
44
44
  scholarx/skills/scholarx-operations/SKILL.md
@@ -7,9 +7,6 @@ scholarx = scholarx.ontology
7
7
  [agent_connector_sdk.skill_providers]
8
8
  scholarx = scholarx.skills
9
9
 
10
- [agent_connector_sdk.source_connector_providers]
11
- scholarx = scholarx.connectors
12
-
13
10
  [console_scripts]
14
11
  scholarx = scholarx.cli:cli
15
12
  scholarx-mcp = scholarx.mcp_server:mcp_server
@@ -1,4 +1,4 @@
1
- agent-connector-sdk[mcp]<1.0.0,>=0.3.18
1
+ agent-connector-sdk[mcp]<1.0.0,>=0.4.1
2
2
  httpx>=0.28.1
3
3
  python-dotenv>=1.2.4
4
4
  Levenshtein>=0.27.5
@@ -10,7 +10,7 @@ defusedxml>=0.7.1
10
10
  scholarx[mcp]
11
11
 
12
12
  [mcp]
13
- agent-connector-sdk[mcp]<1.0.0,>=0.3.18
13
+ agent-connector-sdk[mcp]<1.0.0,>=0.4.1
14
14
 
15
15
  [test]
16
16
  pytest-xdist>=3.8.0
@@ -1,11 +1,11 @@
1
1
  import argparse
2
- import asyncio
3
2
  import json
4
3
  import subprocess
5
4
  from pathlib import Path
6
5
  from unittest.mock import AsyncMock, MagicMock, patch
7
6
 
8
7
  import pytest
8
+
9
9
  from scholarx.cli import (
10
10
  _run_auto_analysis,
11
11
  _show_status,
@@ -427,7 +427,9 @@ def test_run_auto_analysis_success_details(mock_subproc, tmp_path):
427
427
  mock_subproc.side_effect = Exception("Subprocess failure")
428
428
  with patch("pathlib.Path.exists", mock_exists), patch("scholarx.cli.console.print") as mock_print:
429
429
  _run_auto_analysis(args)
430
- any_err = any("Paper analysis failed: Exception" in str(call[0][0]) for call in mock_print.call_args_list if call[0])
430
+ any_err = any(
431
+ "Paper analysis failed: Exception" in str(call[0][0]) for call in mock_print.call_args_list if call[0]
432
+ )
431
433
  assert any_err
432
434
 
433
435
 
@@ -73,7 +73,7 @@ async def test_ingest_entities_writes_nodes_and_edges(ingest):
73
73
  assert set(by_id) == {"a", "b"}
74
74
  assert request.relationships[0].source.record_id == "a"
75
75
  assert request.relationships[0].target.record_id == "b"
76
- assert request.relationships[0].relation_reference.endswith("/relations/publishedInSource")
76
+ assert request.relationships[0].relation_reference.endswith("/publishedInSource")
77
77
 
78
78
 
79
79
  def test_paper_entities_maps_paper_source_category_author():
@@ -371,9 +371,7 @@ async def test_sx_storage_download_url_rejects_traversal(mock_client):
371
371
  mock_client.storage.download_arxiv_id = AsyncMock()
372
372
 
373
373
  with patch("scholarx.mcp_server._get_client", return_value=mock_client):
374
- result = await call_sx_storage(
375
- action="download_url", paper_ids="../../outside,https://example.com/a.pdf"
376
- )
374
+ result = await call_sx_storage(action="download_url", paper_ids="../../outside,https://example.com/a.pdf")
377
375
 
378
376
  assert [item["status"] for item in result["results"]] == ["rejected", "rejected"]
379
377
  mock_client.storage.download_arxiv_id.assert_not_awaited()
@@ -491,9 +489,7 @@ async def test_sx_storage_download_url_deadline_exceeded_precheck(mock_client):
491
489
  patch("scholarx.mcp_server._get_client", return_value=mock_client),
492
490
  patch("scholarx.mcp_server._MAX_DIRECT_DOWNLOAD_SECONDS", -1.0),
493
491
  ):
494
- res = await call_sx_storage(
495
- action="download_url", paper_ids="2603.09022,2603.09023"
496
- )
492
+ res = await call_sx_storage(action="download_url", paper_ids="2603.09022,2603.09023")
497
493
  results = res["results"]
498
494
  assert len(results) == 1
499
495
  assert results[0] == {
@@ -1,8 +1,5 @@
1
- """CONN-SCHOLARX-API-COVERAGE-R001: one MCP tool per remaining API operation, coverage at the 90 baseline."""
1
+ """CONN-SCHOLARX-API-COVERAGE-R001: direct tools for uncovered operations at the measured baseline."""
2
2
 
3
- import subprocess
4
- import sys
5
- from pathlib import Path
6
3
  from unittest.mock import AsyncMock, MagicMock, patch
7
4
 
8
5
  import pytest
@@ -12,8 +9,6 @@ from scholarx.models import PaperSource
12
9
 
13
10
  pytestmark = pytest.mark.spec("CONN-SCHOLARX-API-COVERAGE-R001")
14
11
 
15
- ROOT = Path(__file__).resolve().parents[1]
16
-
17
12
 
18
13
  class MockMCP:
19
14
  def __init__(self):
@@ -86,24 +81,3 @@ async def test_download_urls_requires_ids_and_maps_limit_errors(tools):
86
81
  with patch("scholarx.mcp_server._get_client", return_value=client):
87
82
  assert "error" in await tools.tools["sx_download_urls"](paper_ids="", concurrency=0)
88
83
  assert "error" in await tools.tools["sx_download_urls"](paper_ids="1", concurrency=0)
89
-
90
-
91
- def test_coverage_gate_holds_at_the_baseline():
92
- script = (ROOT / "scripts/verify_api_integration.py").read_text()
93
- assert '"scholarx": 90.0' in script
94
- # The checker names the agent from the directory; run it from a directory called scholarx.
95
- result = subprocess.run(
96
- [
97
- sys.executable,
98
- "-c",
99
- "import sys, importlib.util as u;"
100
- "s=u.spec_from_file_location('v', sys.argv[1]);m=u.module_from_spec(s);s.loader.exec_module(m);"
101
- "r=m.verify_agent(sys.argv[2]);print(r['coverage'])",
102
- str(ROOT / "scripts/verify_api_integration.py"),
103
- str(ROOT),
104
- ],
105
- capture_output=True,
106
- text=True,
107
- check=True,
108
- )
109
- assert float(result.stdout.strip()) >= 90.0
@@ -161,9 +161,7 @@ async def test_download_urls_caps_in_flight(tmp_path):
161
161
  async def download(pid):
162
162
  async with lock:
163
163
  state["in_flight"] += 1
164
- state["max_in_flight"] = max(
165
- state["max_in_flight"], state["in_flight"]
166
- )
164
+ state["max_in_flight"] = max(state["max_in_flight"], state["in_flight"])
167
165
  try:
168
166
  await asyncio.sleep(0.02)
169
167
  path = tmp_path / f"{pid}.pdf"
@@ -204,9 +202,7 @@ async def test_download_urls_skips_existing(tmp_path):
204
202
 
205
203
  client = _client_with_mock_storage(tmp_path)
206
204
  client.storage.download_arxiv_id = download
207
- results = await client.download_urls(
208
- ["https://arxiv.org/abs/2603.00001", "2603.00002"]
209
- )
205
+ results = await client.download_urls(["https://arxiv.org/abs/2603.00001", "2603.00002"])
210
206
  by_id = {r["paper_id"]: r for r in results}
211
207
 
212
208
  assert by_id["2603.00001"]["status"] == "cached"
@@ -1,11 +1,13 @@
1
1
  import asyncio
2
- import pytest
3
2
  import threading
4
3
  import time
5
- from unittest.mock import AsyncMock, MagicMock, patch
6
4
  from pathlib import Path
5
+ from unittest.mock import AsyncMock, MagicMock, patch
6
+
7
+ import pytest
8
+
7
9
  from scholarx.models import Paper, PaperSource
8
- from scholarx.queue import JOB_QUEUE, BACKGROUND_DOWNLOADS, POOL_SIZE
10
+ from scholarx.queue import BACKGROUND_DOWNLOADS, JOB_QUEUE, POOL_SIZE
9
11
 
10
12
 
11
13
  @pytest.fixture(autouse=True)
@@ -143,9 +145,7 @@ def test_queue_pool_drains_many_jobs_with_bounded_concurrency(monkeypatch):
143
145
 
144
146
  job_ids = []
145
147
  for i in range(n_jobs):
146
- paper = Paper(
147
- id=f"arxiv:pool-{i}", title=f"Pool Paper {i}", authors=["Z"], source=PaperSource.ARXIV
148
- )
148
+ paper = Paper(id=f"arxiv:pool-{i}", title=f"Pool Paper {i}", authors=["Z"], source=PaperSource.ARXIV)
149
149
  job_id = f"job-pool-{i}"
150
150
  job_ids.append(job_id)
151
151
  BACKGROUND_DOWNLOADS[job_id] = {"status": "queued", "paper_id": paper.id}
@@ -190,6 +190,6 @@ def test_rate_gate_throttles_starts(monkeypatch):
190
190
  t.join()
191
191
 
192
192
  starts.sort()
193
- gaps = [b - a for a, b in zip(starts, starts[1:])]
193
+ gaps = [b - a for a, b in zip(starts, starts[1:], strict=False)]
194
194
  # Every consecutive pair of starts is separated by ~>= the min interval.
195
195
  assert all(g >= q._MIN_START_INTERVAL - 0.01 for g in gaps)
@@ -1,14 +0,0 @@
1
- {
2
- "_comment": "Tier-1 mcp_tool source preset(s) for scholarx (AU-KG.ingest.mcp-tool-connector). Syncs research papers into the KG as research_paper documents via the scholarx MCP server's sx_search tool. Contributed presets win over the central MCP_TOOL_PRESETS.",
3
- "scholarx-papers": {
4
- "server": "scholarx-mcp",
5
- "tool": "scholarx_search",
6
- "records_path": "papers",
7
- "id_field": "id",
8
- "title_field": "title",
9
- "text_field": "abstract",
10
- "updated_field": "published_date",
11
- "doc_type": "research_paper",
12
- "params_style": "args"
13
- }
14
- }
@@ -1,8 +0,0 @@
1
- {
2
- "algorithm": "agent-connector-sdk:mcp-tool-contract-compat:v2",
3
- "connector": "scholarx-mcp",
4
- "schema_version": "1",
5
- "tools": {
6
- "scholarx_search": "588f116d74c2358f98dd975109b4fae33bd410d805dff3dc287f564924f41e16"
7
- }
8
- }
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes