llama-index-tools-flashdata 0.1.0__tar.gz

llama_index_tools_flashdata-0.1.0/.gitignore

@@ -0,0 +1,7 @@
+.venv/
+__pycache__/
+*.py[cod]
+.pytest_cache/
+.ruff_cache/
+dist/
+*.egg-info/

llama_index_tools_flashdata-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 FlashData contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

llama_index_tools_flashdata-0.1.0/PKG-INFO

@@ -0,0 +1,170 @@
+Metadata-Version: 2.5
+Name: llama-index-tools-flashdata
+Version: 0.1.0
+Summary: Google and YouTube search tools and citation-aware readers for LlamaIndex
+Project-URL: Homepage, https://flashdata.dev
+Project-URL: Documentation, https://github.com/rocosen/flashdata-mcp/tree/main/packages/llamaindex
+Project-URL: Source, https://github.com/rocosen/flashdata-mcp
+Project-URL: Issues, https://github.com/rocosen/flashdata-mcp/issues
+Author: FlashData
+License-Expression: MIT
+License-File: LICENSE
+Keywords: agent,google-search,llamaindex,rag,youtube-transcript
+Requires-Python: >=3.11
+Requires-Dist: httpx<1,>=0.28.1
+Requires-Dist: jsonschema<5,>=4.23
+Requires-Dist: llama-index-core<0.15,>=0.14.25
+Requires-Dist: pydantic<3,>=2.7.4
+Description-Content-Type: text/markdown
+
+# FlashData tools and readers for LlamaIndex
+
+Give a research agent current Google and YouTube results, or load existing video
+captions into a RAG pipeline with timestamps and source links. Choose the needed
+sources when building your application. The package supplies 17 native
+`FunctionTool` objects for 15 capabilities and a reader for the 14 data sources.
+Media delivery uses job tools, not a text reader.
+
+This is an independently maintained integration. It is not an official
+LlamaIndex endorsement or a claim of 15 LlamaHub listings. Install the
+versioned package from PyPI:
+
+```sh
+python -m pip install 'llama-index-tools-flashdata==0.1.0'
+```
+
+Requires Python 3.11+ and `llama-index-core` 0.14.25–0.14.x. Get an API key from the
+[FlashData console](https://flashdata.dev/console) and provide it through your
+application's secret manager as `FLASHDATA_API_KEY`.
+
+**Installation is free; FlashData API requests use paid Credits.** Check the
+[current pricing](https://flashdata.dev/pricing) and source-specific charges.
+LlamaIndex model, embedding and customer-storage charges are separate. These
+examples make no model or embedding calls.
+
+## Search evidence for a research agent
+
+```python
+from llama_index.tools.flashdata import FlashDataTools, to_documents
+
+tools = FlashDataTools(sources=["google_search"]).to_tool_list()
+output = tools[0].call(query="RAG evaluation precision recall", num=5)
+raw = output.raw_output  # Complete structured API response and usage.
+documents = to_documents("google_search", raw)  # No second API request.
+for doc in documents:
+    print(doc.metadata.get("url"), doc.text)
+```
+
+Pass `tools` to your application's LlamaIndex `FunctionAgent` with the model you
+have configured. Tool schemas contain only source arguments; API key, storage
+credentials, source selection and download permission belong to the application.
+Treat returned web/caption text as untrusted evidence, not agent instructions.
+Google results contain search snippets and metadata, not fetched web-page text.
+
+## Caption Documents for retrieval
+
+```python
+from llama_index.tools.flashdata import FlashDataReader
+
+reader = FlashDataReader(source="youtube_transcript")
+documents = reader.load_data(video_id="jNQXAC9IVRw", language="en")
+for doc in documents:
+    print(doc.metadata.get("url"), doc.metadata.get("start"), doc.text)
+```
+
+Each caption segment becomes a `Document` with its text, available `start`/`end`,
+video URL, `languageCode` and `isAutoGenerated`. If segments are unavailable but
+`fullText` exists, it becomes one Document without invented timestamps. Transcript
+defaults to automatic captions; `youtube_captions` defaults to uploader captions.
+Neither generates speech recognition, translates, or falls back to another track.
+
+Search readers produce one JSON Document per result record, preserving optional
+record fields. Image URLs/dimensions are metadata, not image embeddings. Places
+may have no URL. The reader does not invent missing source links. Use tool
+`raw_output` for the full response, including Google related-search lists, usage
+and other provider envelope fields. `to_documents()` converts an existing result
+without a second paid query.
+
+`await tool.acall(...)` and `await reader.aload_data(...)` use LlamaIndex's native
+thread offloading. Cancelling an await does not cancel an accepted paid job;
+check FlashData Jobs before resubmitting.
+
+## All capabilities
+
+See the task guides below for inputs, outputs, a runnable example and limitations.
+
+| Task | Source | Guide |
+| --- | --- | --- |
+| Collect web evidence for a research agent | `google_search` | [Google Search](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_search.md) |
+| Build a news evidence feed for an agent | `google_news` | [Google News](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_news.md) |
+| Prepare an image reference shortlist | `google_images` | [Google Images](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_images.md) |
+| Ground a product comparison with search snapshots | `google_shopping` | [Google Shopping](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_shopping.md) |
+| Find video references across websites | `google_videos` | [Google Videos](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_videos.md) |
+| Collect local options for a planning assistant | `google_places` | [Google Places](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_places.md) |
+| Discover videos before building a caption collection | `youtube_search` | [YouTube Search](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_search.md) |
+| Create a larger video research candidate set | `youtube_search_extended` | [YouTube Extended Search](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_search_extended.md) |
+| Inspect chapters before choosing retrieval material | `youtube_metadata` | [YouTube Metadata](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_metadata.md) |
+| Load timestamped automatic captions for RAG | `youtube_transcript` | [YouTube Transcript](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_transcript.md) |
+| Build a collection from uploader captions | `youtube_captions` | [YouTube Captions](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_captions.md) |
+| Record license-label metadata during dataset review | `youtube_trainability` | [YouTube License Metadata](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_trainability.md) |
+| Snapshot a channel for research triage | `youtube_channel` | [YouTube Channel](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_channel.md) |
+| Explore query phrasing before a video search | `youtube_suggestions` | [YouTube Autocomplete](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_suggestions.md) |
+| Deliver approved media for a downstream workflow | `youtube_download` | [YouTube Download to Cloud Storage](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_download.md) |
+
+## Media delivery lifecycle
+
+Configure `FLASHDATA_DOWNLOAD_STORAGE` as a secret JSON destination. Supported
+providers are S3, S3-compatible, GCS, Azure, OSS, COS, BOS and TOS. Use fields from
+the [download contract](https://flashdata.dev/docs/sources/youtube_download);
+the packaged `storage.schema.json` describes the credential structure. Bucket
+names belong in `storage_url`; optional folders belong in `object_prefix`.
+
+```python
+from llama_index.tools.flashdata import FlashDataTools
+
+# Only enable after authorizing this paid write to your own destination.
+submit, status, result = FlashDataTools(
+    sources=["youtube_download"], allow_download=True
+).to_tool_list()
+receipt = submit.call(video_id="jNQXAC9IVRw", download_type="audio", output_format="m4a").raw_output
+job_id = receipt["id"]  # Persist before any follow-up; never resubmit to poll.
+state = status.call(job_id=job_id).raw_output
+# Later, read result with the same ID. Pending jobs return status, not an error.
+delivery = result.call(job_id=job_id).raw_output
+```
+
+Download submission is disabled by default. You can use status/result tools
+without enabling it or supplying destination credentials. Completed result JSON
+expires after 24 hours; media retention depends on your storage. Supply only
+content you are entitled to download and use. Storage authorization is not a
+training license.
+
+## Failure handling
+
+Calls are single-attempt, bounded to 120 seconds (15-second connection timeout)
+and 20 MiB. Failed/ambiguous POSTs are never automatically retried. A timeout can
+mean the paid job was accepted: check the FlashData Jobs console before another
+submission. Missing credits, unauthorized sources and unavailable captions are
+errors, not successful empty results or reasons to silently change source.
+
+Job reads verify the ID and `youtube_download` source. Download status/receipt
+outputs exclude storage parameters. Result JSON is fetched only from the trusted
+HTTPS origin returned by FlashData authentication; the API key is not sent there.
+Redirects are not followed automatically. Raw HTTP errors and credentials are
+not included in `FlashDataError` messages.
+
+## Local verification
+
+```sh
+uv sync --frozen
+uv run --frozen pytest -q
+uv run --frozen python -m build
+uv run --frozen twine check dist/*
+uv run --frozen python scripts/verify_install.py
+```
+
+Tests disable network sockets and exercise real LlamaIndex tools/readers against
+HTTP fixtures. They do not establish live API, model, storage-write or public
+directory acceptance. Shared parameter and HTTP contracts come from the reviewed
+FlashData Haystack adapter; `contract-provenance.json` records exact origins and
+differences. No Haystack runtime dependency is needed.

llama_index_tools_flashdata-0.1.0/PUBLISHING.md

@@ -0,0 +1,36 @@
+# Publishing the independent package
+
+The build, tests and source installation are separate from PyPI publication
+and official LlamaIndex directory admission. Confirm the actual project/version
+status on PyPI before registering a publisher or dispatching a release.
+
+Use the existing `flashdata-dev` PyPI account. The initial release requires
+owner approval of this project-specific Trusted Publisher; check for an existing
+binding before adding it:
+
+| Field | Value |
+| --- | --- |
+| PyPI project | `llama-index-tools-flashdata` |
+| GitHub owner | `rocosen` |
+| Repository | `flashdata-mcp` |
+| Workflow filename | `llamaindex-publish.yml` |
+| GitHub environment | `pypi-llamaindex` |
+
+Confirm the project is still available before registering. Configure the GitHub
+environment under this repository with deployment restricted to tags matching
+`llamaindex-v*`. The checked-in workflow does not configure either platform.
+
+The workflow runs only by explicit manual dispatch on a `llamaindex-vVERSION`
+tag matching `pyproject.toml`. A normal push, pull request or tag creation does
+not publish the package. It rebuilds, tests and checks the distributions before
+the OIDC publish job. No long-lived PyPI token is required.
+
+Prepare the release README and absolute task-guide links before building,
+because the uploaded distribution contains the PyPI project description. After
+publication, read back the exact version and artifact hashes and install it in a
+fresh environment before marking the release complete. PyPI publication is one
+shared package, not 15 official LlamaIndex/LlamaHub entries.
+
+LlamaIndex no longer accepts new integration packages in its monorepo. Do not
+submit this `pyproject.toml` there. Documentation contributions remain possible,
+but independent listing admission must be established separately.

llama_index_tools_flashdata-0.1.0/README.md

@@ -0,0 +1,151 @@
+# FlashData tools and readers for LlamaIndex
+
+Give a research agent current Google and YouTube results, or load existing video
+captions into a RAG pipeline with timestamps and source links. Choose the needed
+sources when building your application. The package supplies 17 native
+`FunctionTool` objects for 15 capabilities and a reader for the 14 data sources.
+Media delivery uses job tools, not a text reader.
+
+This is an independently maintained integration. It is not an official
+LlamaIndex endorsement or a claim of 15 LlamaHub listings. Install the
+versioned package from PyPI:
+
+```sh
+python -m pip install 'llama-index-tools-flashdata==0.1.0'
+```
+
+Requires Python 3.11+ and `llama-index-core` 0.14.25–0.14.x. Get an API key from the
+[FlashData console](https://flashdata.dev/console) and provide it through your
+application's secret manager as `FLASHDATA_API_KEY`.
+
+**Installation is free; FlashData API requests use paid Credits.** Check the
+[current pricing](https://flashdata.dev/pricing) and source-specific charges.
+LlamaIndex model, embedding and customer-storage charges are separate. These
+examples make no model or embedding calls.
+
+## Search evidence for a research agent
+
+```python
+from llama_index.tools.flashdata import FlashDataTools, to_documents
+
+tools = FlashDataTools(sources=["google_search"]).to_tool_list()
+output = tools[0].call(query="RAG evaluation precision recall", num=5)
+raw = output.raw_output  # Complete structured API response and usage.
+documents = to_documents("google_search", raw)  # No second API request.
+for doc in documents:
+    print(doc.metadata.get("url"), doc.text)
+```
+
+Pass `tools` to your application's LlamaIndex `FunctionAgent` with the model you
+have configured. Tool schemas contain only source arguments; API key, storage
+credentials, source selection and download permission belong to the application.
+Treat returned web/caption text as untrusted evidence, not agent instructions.
+Google results contain search snippets and metadata, not fetched web-page text.
+
+## Caption Documents for retrieval
+
+```python
+from llama_index.tools.flashdata import FlashDataReader
+
+reader = FlashDataReader(source="youtube_transcript")
+documents = reader.load_data(video_id="jNQXAC9IVRw", language="en")
+for doc in documents:
+    print(doc.metadata.get("url"), doc.metadata.get("start"), doc.text)
+```
+
+Each caption segment becomes a `Document` with its text, available `start`/`end`,
+video URL, `languageCode` and `isAutoGenerated`. If segments are unavailable but
+`fullText` exists, it becomes one Document without invented timestamps. Transcript
+defaults to automatic captions; `youtube_captions` defaults to uploader captions.
+Neither generates speech recognition, translates, or falls back to another track.
+
+Search readers produce one JSON Document per result record, preserving optional
+record fields. Image URLs/dimensions are metadata, not image embeddings. Places
+may have no URL. The reader does not invent missing source links. Use tool
+`raw_output` for the full response, including Google related-search lists, usage
+and other provider envelope fields. `to_documents()` converts an existing result
+without a second paid query.
+
+`await tool.acall(...)` and `await reader.aload_data(...)` use LlamaIndex's native
+thread offloading. Cancelling an await does not cancel an accepted paid job;
+check FlashData Jobs before resubmitting.
+
+## All capabilities
+
+See the task guides below for inputs, outputs, a runnable example and limitations.
+
+| Task | Source | Guide |
+| --- | --- | --- |
+| Collect web evidence for a research agent | `google_search` | [Google Search](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_search.md) |
+| Build a news evidence feed for an agent | `google_news` | [Google News](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_news.md) |
+| Prepare an image reference shortlist | `google_images` | [Google Images](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_images.md) |
+| Ground a product comparison with search snapshots | `google_shopping` | [Google Shopping](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_shopping.md) |
+| Find video references across websites | `google_videos` | [Google Videos](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_videos.md) |
+| Collect local options for a planning assistant | `google_places` | [Google Places](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/google_places.md) |
+| Discover videos before building a caption collection | `youtube_search` | [YouTube Search](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_search.md) |
+| Create a larger video research candidate set | `youtube_search_extended` | [YouTube Extended Search](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_search_extended.md) |
+| Inspect chapters before choosing retrieval material | `youtube_metadata` | [YouTube Metadata](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_metadata.md) |
+| Load timestamped automatic captions for RAG | `youtube_transcript` | [YouTube Transcript](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_transcript.md) |
+| Build a collection from uploader captions | `youtube_captions` | [YouTube Captions](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_captions.md) |
+| Record license-label metadata during dataset review | `youtube_trainability` | [YouTube License Metadata](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_trainability.md) |
+| Snapshot a channel for research triage | `youtube_channel` | [YouTube Channel](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_channel.md) |
+| Explore query phrasing before a video search | `youtube_suggestions` | [YouTube Autocomplete](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_suggestions.md) |
+| Deliver approved media for a downstream workflow | `youtube_download` | [YouTube Download to Cloud Storage](https://github.com/rocosen/flashdata-mcp/blob/llamaindex-v0.1.0/packages/llamaindex/guides/youtube_download.md) |
+
+## Media delivery lifecycle
+
+Configure `FLASHDATA_DOWNLOAD_STORAGE` as a secret JSON destination. Supported
+providers are S3, S3-compatible, GCS, Azure, OSS, COS, BOS and TOS. Use fields from
+the [download contract](https://flashdata.dev/docs/sources/youtube_download);
+the packaged `storage.schema.json` describes the credential structure. Bucket
+names belong in `storage_url`; optional folders belong in `object_prefix`.
+
+```python
+from llama_index.tools.flashdata import FlashDataTools
+
+# Only enable after authorizing this paid write to your own destination.
+submit, status, result = FlashDataTools(
+    sources=["youtube_download"], allow_download=True
+).to_tool_list()
+receipt = submit.call(video_id="jNQXAC9IVRw", download_type="audio", output_format="m4a").raw_output
+job_id = receipt["id"]  # Persist before any follow-up; never resubmit to poll.
+state = status.call(job_id=job_id).raw_output
+# Later, read result with the same ID. Pending jobs return status, not an error.
+delivery = result.call(job_id=job_id).raw_output
+```
+
+Download submission is disabled by default. You can use status/result tools
+without enabling it or supplying destination credentials. Completed result JSON
+expires after 24 hours; media retention depends on your storage. Supply only
+content you are entitled to download and use. Storage authorization is not a
+training license.
+
+## Failure handling
+
+Calls are single-attempt, bounded to 120 seconds (15-second connection timeout)
+and 20 MiB. Failed/ambiguous POSTs are never automatically retried. A timeout can
+mean the paid job was accepted: check the FlashData Jobs console before another
+submission. Missing credits, unauthorized sources and unavailable captions are
+errors, not successful empty results or reasons to silently change source.
+
+Job reads verify the ID and `youtube_download` source. Download status/receipt
+outputs exclude storage parameters. Result JSON is fetched only from the trusted
+HTTPS origin returned by FlashData authentication; the API key is not sent there.
+Redirects are not followed automatically. Raw HTTP errors and credentials are
+not included in `FlashDataError` messages.
+
+## Local verification
+
+```sh
+uv sync --frozen
+uv run --frozen pytest -q
+uv run --frozen python -m build
+uv run --frozen twine check dist/*
+uv run --frozen python scripts/verify_install.py
+```
+
+Tests disable network sockets and exercise real LlamaIndex tools/readers against
+HTTP fixtures. They do not establish live API, model, storage-write or public
+directory acceptance. Shared parameter and HTTP contracts come from the reviewed
+FlashData Haystack adapter; `contract-provenance.json` records exact origins and
+differences. No Haystack runtime dependency is needed.

llama_index_tools_flashdata-0.1.0/capabilities.json

@@ -0,0 +1,171 @@
+[
+  {
+    "source": "google_search",
+    "name": "Google Search",
+    "sample": {
+      "query": "retrieval augmented generation evaluation",
+      "num": 5
+    },
+    "inputs": "query; optional gl, hl, location, num/page 1–100, tbs, autocorrect",
+    "outputs": "results[].organic: title, link, snippet, position; usage.credits",
+    "limits": "Search snippets are not fetched page content. Omitted count and locale are not filled in by the SDK."
+  },
+  {
+    "source": "google_news",
+    "name": "Google News",
+    "sample": {
+      "query": "reusable rockets",
+      "num": 5
+    },
+    "inputs": "query; optional Google search parameters",
+    "outputs": "results[].news: titles, links and available dates/snippets; usage.credits",
+    "limits": "Returns news search results, not article full text."
+  },
+  {
+    "source": "google_images",
+    "name": "Google Images",
+    "sample": {
+      "query": "rooftop solar installation",
+      "num": 5
+    },
+    "inputs": "query; optional Google search parameters",
+    "outputs": "results[].images: image/source URLs and available dimensions; usage.credits",
+    "limits": "Does not download images or establish image reuse rights."
+  },
+  {
+    "source": "google_shopping",
+    "name": "Google Shopping",
+    "sample": {
+      "query": "USB microphone",
+      "num": 5
+    },
+    "inputs": "query; optional Google search parameters",
+    "outputs": "results[].shopping: title, link, price and available merchant/rating fields; usage.credits",
+    "limits": "Prices are search snapshots; inventory and checkout price are not guaranteed."
+  },
+  {
+    "source": "google_videos",
+    "name": "Google Videos",
+    "sample": {
+      "query": "database indexing tutorial",
+      "num": 5
+    },
+    "inputs": "query; optional Google search parameters",
+    "outputs": "results[].videos: video titles, links and available metadata; usage.credits",
+    "limits": "Search discovers videos; it does not return video files or transcripts."
+  },
+  {
+    "source": "google_places",
+    "name": "Google Places",
+    "sample": {
+      "query": "coworking Singapore",
+      "num": 5
+    },
+    "inputs": "query; optional location and other Google search parameters",
+    "outputs": "results[].places: names and available addresses, ratings and contact fields; usage.credits",
+    "limits": "Contact details may be missing; verify current hours and availability with the business."
+  },
+  {
+    "source": "youtube_search",
+    "name": "YouTube Search",
+    "sample": {
+      "query": "RAG evaluation tutorials",
+      "max_results": 5
+    },
+    "inputs": "query; max_results 1–20 (default 10); optional filters",
+    "outputs": "results[].results: video search records; totalResults and query; usage.credits",
+    "limits": "The requested count is a ceiling, not a guaranteed result count."
+  },
+  {
+    "source": "youtube_search_extended",
+    "name": "YouTube Extended Search",
+    "sample": {
+      "query": "retrieval evaluation tutorials",
+      "max_results": 50
+    },
+    "inputs": "query; max_results 1–700 (default 20); optional filters",
+    "outputs": "results[].results: video search records; totalResults and query; usage.credits",
+    "limits": "This is a distinct source with a separate higher rate than standard search. Requested count is not guaranteed."
+  },
+  {
+    "source": "youtube_metadata",
+    "name": "YouTube Metadata",
+    "sample": {
+      "video_id": "dQw4w9WgXcQ",
+      "include_formats": false,
+      "include_chapters": true
+    },
+    "inputs": "video_id (11 characters); include_formats/include_chapters (default true)",
+    "outputs": "results[]: videoId, title, channel, duration and other available metadata; usage.credits",
+    "limits": "Metadata is not media download. Chapters and optional fields may be absent."
+  },
+  {
+    "source": "youtube_transcript",
+    "name": "YouTube Transcript",
+    "sample": {
+      "video_id": "dQw4w9WgXcQ",
+      "language": "en",
+      "auto": true
+    },
+    "inputs": "video_id; language (default en); auto (default true)",
+    "outputs": "results[]: videoId, languageCode, isAutoGenerated, segments, fullText; usage.credits",
+    "limits": "Reads an existing track. No ASR, translation, or automatic fallback to uploader captions."
+  },
+  {
+    "source": "youtube_captions",
+    "name": "YouTube Captions",
+    "sample": {
+      "video_id": "dQw4w9WgXcQ",
+      "language": "en",
+      "auto": false
+    },
+    "inputs": "video_id; language (default en); auto (default false)",
+    "outputs": "results[]: videoId, languageCode, isAutoGenerated, segments, fullText; usage.credits",
+    "limits": "Defaults to uploader captions. No ASR, translation, or automatic fallback to generated captions."
+  },
+  {
+    "source": "youtube_trainability",
+    "name": "YouTube License Metadata",
+    "sample": {
+      "video_id": "dQw4w9WgXcQ"
+    },
+    "inputs": "video_id (11 characters)",
+    "outputs": "results[]: videoId and license-label indicator permitted, with available kind/etag; usage.credits",
+    "limits": "permitted is a license-label signal, not legal training permission or creator third-party training settings."
+  },
+  {
+    "source": "youtube_channel",
+    "name": "YouTube Channel",
+    "sample": {
+      "handle": "@GoogleDevelopers",
+      "video_limit": 5
+    },
+    "inputs": "handle; video_limit 1–100 (default 20)",
+    "outputs": "results[]: channelId, channelHandle, name, channelUrl, statistics, videos; usage.credits",
+    "limits": "A recent-video snapshot is not the complete channel archive."
+  },
+  {
+    "source": "youtube_suggestions",
+    "name": "YouTube Autocomplete",
+    "sample": {
+      "query": "vector database",
+      "language": "en",
+      "region": "US"
+    },
+    "inputs": "query; language (default en); region (default US, two letters)",
+    "outputs": "results: autocomplete suggestion strings; usage.credits",
+    "limits": "Suggestions are not keyword volume or popularity measurements."
+  },
+  {
+    "source": "youtube_download",
+    "name": "YouTube Download to Cloud Storage",
+    "sample": {
+      "video_id": "dQw4w9WgXcQ",
+      "download_type": "audio",
+      "output_format": "m4a"
+    },
+    "inputs": "submit: video_id, optional format/quality/clip fields; reads: job_id. Storage comes from FLASHDATA_DOWNLOAD_STORAGE.",
+    "outputs": "submit: id/status receipt; reads: job lifecycle; completed result: output JSON delivery manifest",
+    "limits": "Paid asynchronous job. Requires customer storage. Save the receipt; never submit again to poll. Result JSON expires after 24 hours; media retention depends on storage."
+  }
+]

llama_index_tools_flashdata-0.1.0/contract-provenance.json

@@ -0,0 +1,26 @@
+{
+  "base_commit": "3e22639dd247412380ae5e097d5261202a63df1e",
+  "note": "Reuse Haystack contract. Only HTTP user-agent/docstring and removal of unused legacy Google defaults differ.",
+  "files": {
+    "_client.py": {
+      "source": "packages/haystack/src/flashdata_haystack/_client.py",
+      "source_sha256": "d5abddeb4ecd0944da6226085821fafacd71349d16eea65d5ddb47bb469088e3",
+      "sha256": "54e4fb65e8319e65016f3c7d2e3aa4e1f137fdad0c76bdda6e0e700010157ef1"
+    },
+    "_inputs.py": {
+      "source": "packages/haystack/src/flashdata_haystack/_inputs.py",
+      "source_sha256": "b55ea6429ffdaa765adfedb8d985dd2771847d987d97e967bc532b74710518d6",
+      "sha256": "df9057e47cdb9a017326cd379053a386aa4bbb8862d6a723be91322444451ff1"
+    },
+    "_storage.py": {
+      "source": "packages/haystack/src/flashdata_haystack/_storage.py",
+      "source_sha256": "88e24e19c3844ca3c4e07645e11b011b70e244e106d4416e7b12b5bff2c29724",
+      "sha256": "88e24e19c3844ca3c4e07645e11b011b70e244e106d4416e7b12b5bff2c29724"
+    },
+    "storage.schema.json": {
+      "source": "packages/haystack/src/flashdata_haystack/storage.schema.json",
+      "source_sha256": "a690171fc7d129ec18aa2239effafaadf78b490e292aa2196de581d63cdb88f6",
+      "sha256": "a690171fc7d129ec18aa2239effafaadf78b490e292aa2196de581d63cdb88f6"
+    }
+  }
+}

llama_index_tools_flashdata-0.1.0/examples/google_images.py

@@ -0,0 +1,15 @@
+"""Prepare an image reference shortlist. Running this example makes one paid FlashData request."""
+
+from llama_index.tools.flashdata import FlashDataTools, to_documents
+
+
+def main():
+    tool = FlashDataTools(sources=["google_images"]).to_tool_list()[0]
+    output = tool.call(query="rooftop solar installation", num=5)
+    print(output.raw_output)
+    for document in to_documents("google_images", output.raw_output):
+        print(document.metadata, document.text)
+
+
+if __name__ == "__main__":
+    main()

llama_index_tools_flashdata-0.1.0/examples/google_news.py

@@ -0,0 +1,15 @@
+"""Build a news evidence feed for an agent. Running this example makes one paid FlashData request."""
+
+from llama_index.tools.flashdata import FlashDataTools, to_documents
+
+
+def main():
+    tool = FlashDataTools(sources=["google_news"]).to_tool_list()[0]
+    output = tool.call(query="reusable rockets", num=5)
+    print(output.raw_output)
+    for document in to_documents("google_news", output.raw_output):
+        print(document.metadata, document.text)
+
+
+if __name__ == "__main__":
+    main()

llama_index_tools_flashdata-0.1.0/examples/google_places.py

@@ -0,0 +1,15 @@
+"""Collect local options for a planning assistant. Running this example makes one paid FlashData request."""
+
+from llama_index.tools.flashdata import FlashDataTools, to_documents
+
+
+def main():
+    tool = FlashDataTools(sources=["google_places"]).to_tool_list()[0]
+    output = tool.call(query="coworking Singapore", num=5)
+    print(output.raw_output)
+    for document in to_documents("google_places", output.raw_output):
+        print(document.metadata, document.text)
+
+
+if __name__ == "__main__":
+    main()

llama_index_tools_flashdata-0.1.0/examples/google_search.py

@@ -0,0 +1,15 @@
+"""Collect web evidence for a research agent. Running this example makes one paid FlashData request."""
+
+from llama_index.tools.flashdata import FlashDataTools, to_documents
+
+
+def main():
+    tool = FlashDataTools(sources=["google_search"]).to_tool_list()[0]
+    output = tool.call(query="retrieval augmented generation evaluation", num=5)
+    print(output.raw_output)
+    for document in to_documents("google_search", output.raw_output):
+        print(document.metadata, document.text)
+
+
+if __name__ == "__main__":
+    main()