Add a 32-layer landscape-sweep lane to manifest-20260923 and source reviews for its newcomers

evidence/artifacts/landscape-sweep-20260923/datalab-to-marker.json

@@ -0,0 +1,33 @@
+{
+ "schema_version": 1,
+ "id": "source-review-datalab-to-marker",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/datalab-to/marker",
+ "reviewed_commit": "8a1d2344de25d7ec4c5209133aed7af565874ff0",
+ "readme_path": "README.md",
+ "license": "Apache-2.0",
+ "layers": [
+  "document-retrieval"
+ ],
+ "claim": "Source and documentation review of datalab-to/marker at commit 8a1d2344de25d7ec4c5209133aed7af565874ff0 (license Apache-2.0), read from README.md at that commit. Repository description: \"Convert PDF to markdown + JSON quickly with high accuracy\". Reviewer summary of the upstream documentation: Datalab's open-source pipeline converting PDFs, images, DOCX/PPTX/XLSX, HTML and EPUB to Markdown/JSON/chunks, using its own Surya OCR/layout/table/equation models on GPU, CPU or MPS, with an optional LLM-refinement pass. Requirement part it addresses (document-retrieval): Directly produces 'source-grounded Markdown/document content' through 'controlled ingestion' of PDFs/Office/HTML: per its own README it reads the PDF text layer and OCRs selectively, so extracted text stays grounded in the original page content and reading order, including tables/equations/multi-column layout.",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@8a1d2344de25d7ec4c5209133aed7af565874ff0",
+   "text": "Marker converts documents to markdown, JSON, chunks, and HTML quickly and accurately."
+  },
+  {
+   "source": "README.md@8a1d2344de25d7ec4c5209133aed7af565874ff0",
+   "text": "If you have high volume workloads, we offer a batch processing service that has processed 1B+ pages per week \u2014 we manage the infrastructure so your workloads finish on time."
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/datalab-to/marker/tree/8a1d2344de25d7ec4c5209133aed7af565874ff0",
+  "https://github.com/datalab-to/marker/blob/8a1d2344de25d7ec4c5209133aed7af565874ff0/README.md",
+  "https://jimmysong.io/blog/pdf-to-markdown-open-source-deep-dive/"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/exa-labs-exa-mcp-server.json

@@ -0,0 +1,35 @@
+{
+ "schema_version": 1,
+ "id": "source-review-exa-labs-exa-mcp-server",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/exa-labs/exa-mcp-server",
+ "reviewed_commit": "f3d71fb6b0ff4b4683f108f05bc2bae61a9f7e97",
+ "readme_path": "README.md",
+ "license": "MIT",
+ "layers": [
+  "web-research"
+ ],
+ "claim": "Source and documentation review of exa-labs/exa-mcp-server at commit f3d71fb6b0ff4b4683f108f05bc2bae61a9f7e97 (license MIT), read from README.md at that commit. Repository description: \"Exa MCP for web search and web crawling!\". Reviewer summary of the upstream documentation: Official MCP server from Exa Labs exposing Exa's neural/semantic web search, crawling and entity discovery (people, companies, code, papers) as bounded MCP tool calls; backed by the official exa-labs/exa-js SDK. Requirement part it addresses (web-research): \"Retrieve attributable primary sources ...",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@f3d71fb6b0ff4b4683f108f05bc2bae61a9f7e97",
+   "text": "Connect AI agents to Exa for web search, content fetching, and multi-step research."
+  },
+  {
+   "source": "README.md@f3d71fb6b0ff4b4683f108f05bc2bae61a9f7e97",
+   "text": "Documentation  \u00b7  NPM Package  \u00b7  Get your API key"
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/exa-labs/exa-mcp-server/tree/f3d71fb6b0ff4b4683f108f05bc2bae61a9f7e97",
+  "https://github.com/exa-labs/exa-mcp-server/blob/f3d71fb6b0ff4b4683f108f05bc2bae61a9f7e97/README.md",
+  "https://exa.ai/versus/tavily",
+  "https://www.tavily.com/blog/7-best-exa-alternatives-for-ai-agents-in-2026",
+  "https://medium.com/@unicodeveloper/tavily-alternatives-in-2026-after-the-nebius-acquisition-9de526780686"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/github-gh-aw.json

@@ -0,0 +1,32 @@
+{
+ "schema_version": 1,
+ "id": "source-review-github-gh-aw",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/github/gh-aw",
+ "reviewed_commit": "358fbf270e2a97c5aeb90af1e911a73df8dcc07b",
+ "readme_path": "README.md",
+ "license": "MIT",
+ "layers": [
+  "git-github-automation"
+ ],
+ "claim": "Source and documentation review of github/gh-aw at commit 358fbf270e2a97c5aeb90af1e911a73df8dcc07b (license MIT), read from README.md at that commit. Repository description: \"GitHub Agentic Workflows\". Requirement part it addresses (git-github-automation): GitHub review/PR automation",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@358fbf270e2a97c5aeb90af1e911a73df8dcc07b",
+   "text": "Hello fellow agent! Welcome to GitHub Agentic Workflows = Actions + Agent + Safety. Here are some pointers to get you started in using this tool."
+  },
+  {
+   "source": "README.md@358fbf270e2a97c5aeb90af1e911a73df8dcc07b",
+   "text": "GitHub Agentic Workflows (gh-aw) lets developers define AI-powered repository automation in Markdown with YAML frontmatter and run AI agents securely through GitHub Actions."
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/github/gh-aw/tree/358fbf270e2a97c5aeb90af1e911a73df8dcc07b",
+  "https://github.com/github/gh-aw/blob/358fbf270e2a97c5aeb90af1e911a73df8dcc07b/README.md"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/google-osv-scanner.json

@@ -0,0 +1,32 @@
+{
+ "schema_version": 1,
+ "id": "source-review-google-osv-scanner",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/google/osv-scanner",
+ "reviewed_commit": "47f4ca0c48952ce9672ec81b0d03f328a3ea6733",
+ "readme_path": "README.md",
+ "license": "Apache-2.0",
+ "layers": [
+  "security-supply-chain"
+ ],
+ "claim": "Source and documentation review of google/osv-scanner at commit 47f4ca0c48952ce9672ec81b0d03f328a3ea6733 (license Apache-2.0), read from README.md at that commit. Repository description: \"Vulnerability scanner written in Go which uses the data provided by https://osv.dev\". Reviewer summary of the upstream documentation: Google/OSV.dev's official vulnerability scanner.",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@47f4ca0c48952ce9672ec81b0d03f328a3ea6733",
+   "text": "Use OSV-Scanner to find existing vulnerabilities affecting your project's dependencies. OSV-Scanner provides an officially supported frontend to the OSV database and CLI interface to OSV-Scalibr that connects a project\u2019s list of dependencies with the vulnerabilities that affect them."
+  },
+  {
+   "source": "README.md@47f4ca0c48952ce9672ec81b0d03f328a3ea6733",
+   "text": "OSV-Scanner supports a wide range of project types, package managers and features, including but not limited to:"
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/google/osv-scanner/tree/47f4ca0c48952ce9672ec81b0d03f328a3ea6733",
+  "https://github.com/google/osv-scanner/blob/47f4ca0c48952ce9672ec81b0d03f328a3ea6733/README.md"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/grafana-tempo.json

@@ -0,0 +1,35 @@
+{
+ "schema_version": 1,
+ "id": "source-review-grafana-tempo",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/grafana/tempo",
+ "reviewed_commit": "e2c16d6e4d7e03120b31f56ccb0ed3d5788db91f",
+ "readme_path": "README.md",
+ "license": "AGPL-3.0",
+ "layers": [
+  "observation-inference"
+ ],
+ "claim": "Source and documentation review of grafana/tempo at commit e2c16d6e4d7e03120b31f56ccb0ed3d5788db91f (license AGPL-3.0), read from README.md at that commit. Repository description: \"Grafana Tempo is a high volume, minimal dependency distributed tracing backend.\". Reviewer summary of the upstream documentation: Grafana Labs' open-source, object-storage-backed distributed tracing backend. Requirement part it addresses (observation-inference): Requirement: 'Observe real native usage/service outcomes through one privacy-filtered local pipeline...'. Tempo is the trace-signal component of that pipeline;",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@e2c16d6e4d7e03120b31f56ccb0ed3d5788db91f",
+   "text": "Grafana Tempo is an open source, easy-to-use, and high-scale distributed tracing backend."
+  },
+  {
+   "source": "README.md@e2c16d6e4d7e03120b31f56ccb0ed3d5788db91f",
+   "text": "Distributed tracing helps teams quickly pinpoint performance issues and understand the flow of requests across services. The Traces Drilldown UI simplifies this process by offering a user-friendly interface to view and analyze trace data, making it easier to identify and resolve issues without needing to write complex queries."
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/grafana/tempo/tree/e2c16d6e4d7e03120b31f56ccb0ed3d5788db91f",
+  "https://github.com/grafana/tempo/blob/e2c16d6e4d7e03120b31f56ccb0ed3d5788db91f/README.md",
+  "https://grafana.com/oss/tempo/",
+  "https://signoz.io/blog/jaeger-vs-tempo/",
+  "https://www.dash0.com/comparisons/open-source-distributed-tracing-tools"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/lightgbm-org-lightgbm.json

@@ -0,0 +1,33 @@
+{
+ "schema_version": 1,
+ "id": "source-review-lightgbm-org-lightgbm",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/lightgbm-org/LightGBM",
+ "reviewed_commit": "9149cae965332053e82c8d85afe71374662a3185",
+ "readme_path": "README.md",
+ "license": "MIT",
+ "layers": [
+  "research-factors-ml"
+ ],
+ "claim": "Source and documentation review of lightgbm-org/LightGBM at commit 9149cae965332053e82c8d85afe71374662a3185 (license MIT), read from README.md at that commit. Repository description: \"A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.\". Requirement part it addresses (research-factors-ml): Requirement text: \"...qualify any strategy with causal data, costs and independent reconciliation.\" LightGBM is the concrete model-fitting engine the layer's own known repositories run against the causal, point-in-time factor panel before a strategy can be qualified: Qlib's default benchmark baseline (LGBModel, and the DoubleEnsemble baseline whose base model is LGBM) and stefan-jansen/machine-learning-for-trading's ML-factor chapters both depend on it directly.",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@9149cae965332053e82c8d85afe71374662a3185",
+   "text": "Light Gradient Boosting Machine ==============================="
+  },
+  {
+   "source": "README.md@9149cae965332053e82c8d85afe71374662a3185",
+   "text": "LightGBM is a gradient boosting framework that uses tree based learning algorithms. It is designed to be distributed and efficient with the following advantages:"
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/lightgbm-org/LightGBM/tree/9149cae965332053e82c8d85afe71374662a3185",
+  "https://github.com/lightgbm-org/LightGBM/blob/9149cae965332053e82c8d85afe71374662a3185/README.md",
+  "https://github.com/lightgbm-org/LightGBM"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/microsoft-rd-agent.json

@@ -0,0 +1,33 @@
+{
+ "schema_version": 1,
+ "id": "source-review-microsoft-rd-agent",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/microsoft/RD-Agent",
+ "reviewed_commit": "484776c211e4fbbeef03e0ec00d6bbee7362a4f4",
+ "readme_path": "README.md",
+ "license": "MIT",
+ "layers": [
+  "research-factors-ml"
+ ],
+ "claim": "Source and documentation review of microsoft/RD-Agent at commit 484776c211e4fbbeef03e0ec00d6bbee7362a4f4 (license MIT), read from README.md at that commit. Repository description: \"Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are committed to automating these high-value generic R&D processes through R&D-Agent, which lets AI drive data-driven AI. \\ud83d\\udd17https://aka.ms/RD-Agent-Tech-Report\". Reviewer summary of the upstream documentation: Microsoft's official multi-agent framework for automating research-and-development loops; its RD-Agent(Q) scenario is described in its own README as 'the first data-centric, multi-agent framework designed to automate the full-stack research and development of quantitative strategies via coordinated factor-model co-optimization,' built directly on top of Qlib, with dedicated commands (fin_quant, fin_factor, fin_model, fin_factor_report).",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@484776c211e4fbbeef03e0ec00d6bbee7362a4f4",
+   "text": "Agent\u00b2 RL-Bench evaluates whether LLM agents can autonomously engineer end-to-end post-training pipelines, spanning static optimization and stateful online RL."
+  },
+  {
+   "source": "README.md@484776c211e4fbbeef03e0ec00d6bbee7362a4f4",
+   "text": "MLE-bench is a comprehensive benchmark evaluating the performance of AI agents on machine learning engineering tasks. Utilizing datasets from 75 Kaggle competitions, MLE-bench provides robust assessments of AI systems' capabilities in real-world ML engineering scenarios."
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/microsoft/RD-Agent/tree/484776c211e4fbbeef03e0ec00d6bbee7362a4f4",
+  "https://github.com/microsoft/RD-Agent/blob/484776c211e4fbbeef03e0ec00d6bbee7362a4f4/README.md",
+  "https://github.com/microsoft/RD-Agent"
+ ]
+}

evidence/artifacts/landscape-sweep-20260923/opendatalab-mineru.json

@@ -0,0 +1,32 @@
+{
+ "schema_version": 1,
+ "id": "source-review-opendatalab-mineru",
+ "kind": "upstream_provenance",
+ "evidence_class": "source_review",
+ "repository": "https://github.com/opendatalab/MinerU",
+ "reviewed_commit": "9227669985a7876abe5d3181b1cda11a844e7322",
+ "readme_path": "README.md",
+ "license": null,
+ "layers": [
+  "document-retrieval"
+ ],
+ "claim": "Source and documentation review of opendatalab/MinerU at commit 9227669985a7876abe5d3181b1cda11a844e7322, read from README.md at that commit. Repository description: \"Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.\". Reviewer summary of the upstream documentation: OpenDataLab/Shanghai AI Lab's open-source pipeline transforming complex PDFs and Office documents (multi-column, tables, formulas, CJK) into LLM-ready Markdown/JSON; Requirement part it addresses (document-retrieval): arbitrary filings' left unaddressed.",
+ "documentation_excerpts": [
+  {
+   "source": "README.md@9227669985a7876abe5d3181b1cda11a844e7322",
+   "text": "MinerU 4.0 brings document parsing, a local document library, and service tools into one workflow for document conversion, application integration, and agent reading."
+  },
+  {
+   "source": "README.md@9227669985a7876abe5d3181b1cda11a844e7322",
+   "text": "PDF and images support all four tiers. Office, OpenDocument, EPUB, OFD, HTML/MHTML, and CSV/TSV use local Flash native parsing. Plain text is read directly rather than parsed. Documents are not automatically uploaded to the official service; remote parsing requires explicit configuration."
+  }
+ ],
+ "limitations": [
+  "Source and documentation review at the pinned commit only: not installed, executed, benchmarked or compared with any other component.",
+  "The description and excerpts are the upstream project's own words at that commit; the reviewer summary was written by a research agent and checked by an independent fit review. Both are upstream claims, not measured results."
+ ],
+ "sources": [
+  "https://github.com/opendatalab/MinerU/tree/9227669985a7876abe5d3181b1cda11a844e7322",
+  "https://github.com/opendatalab/MinerU/blob/9227669985a7876abe5d3181b1cda11a844e7322/README.md"
+ ]
+}