flashdata-dev commited on
Commit
43ae5be
·
verified ·
1 Parent(s): 8554644

Clarify capability names and simplify documentation

Browse files
Files changed (4) hide show
  1. README.md +32 -35
  2. index.html +5 -5
  3. task.json +6 -5
  4. tool.py +1 -1
README.md CHANGED
@@ -1,12 +1,13 @@
1
  ---
2
- title: "YouTube transcripts for cited research"
3
- emoji: 🔎
4
- colorFrom: blue
5
- colorTo: indigo
6
- sdk: static
7
- app_file: index.html
 
8
  pinned: false
9
- license: mit
10
  tags:
11
  - smolagents
12
  - tool
@@ -14,25 +15,27 @@ tags:
14
  - youtube-transcript
15
  ---
16
 
17
- # YouTube transcripts for cited research — FlashData
18
 
19
- Give a research agent timestamped text from an existing YouTube caption track. Keep segment start and end times so a reader can return to the video when checking a generated summary.
20
 
21
- **Try this task:** Read existing English automatic captions for a timestamped video brief.
22
 
23
- [FlashData youtube transcript documentation](https://flashdata.dev/docs/sources/youtube_transcript?utm_source=hugging-face&utm_medium=integration&utm_campaign=intent_gtm_2026q4&utm_content=youtube-transcript-research-flashdata) · [Create a FlashData account](https://flashdata.dev/signup)
24
 
25
- ## Run it
26
 
27
- Open the Space to edit a request and inspect a documentation response sample. The interactive page runs locally in your browser; it does not send a FlashData query or collect an API key. The Python tool runs in your own process.
28
 
29
- Use Python 3.10 or newer. Install the pinned dependencies:
 
 
30
 
31
  ```sh
32
  python -m pip install "smolagents==1.26.0" "langchain-flashdata==0.2.1"
33
  ```
34
 
35
- Create an API key in the FlashData console, fund the required credits, and supply the key in your local environment. Review `tool.py` before enabling remote code execution.
36
 
37
  ```python
38
  import getpass
@@ -45,34 +48,28 @@ result = tool(**{'auto': True, 'language': 'en', 'video_id': 'FUfGcZ092b0'})
45
  print(result)
46
  ```
47
 
48
- This is a native smolagents Tool with typed task inputs. Add the loaded tool to your own agent; model choice, prompts and model fees remain under your control. Direct calls above do not need a model.
49
 
50
  ## Inputs
51
 
52
- | Parameter | Type | Required / default | Details |
53
- | --- | --- | --- | --- |
54
- | `video_id` | string | Required | 11-character YouTube video ID or a YouTube video URL. |
55
- | `language` | string | 'en' | Requested caption language code. pattern=^[A-Za-z0-9-]{1,35}$ |
56
- | `auto` | boolean | True | True requests automatic captions; false requests manually authored captions. |
57
-
58
- Optional `None` values are omitted; the documented adapter defaults apply.
59
-
60
- Full validation rules are in `input-schema.json`.
61
-
62
- ## Output and next step
63
 
64
- The useful data is at `results.0.segments`. Relevant fields: `start`, `end`, `text`.
65
 
66
- Retain video ID, language and timestamps when passing segments to your own model. Model calls and summarization are separate steps.
67
 
68
- The Space includes a recorded documentation example and its original request. It is not a fresh result for the task prompt above. Optional fields may be absent.
69
 
70
- ## Costs, limits and failures
71
 
72
- Defaults to existing automatic English captions. No ASR, translation or automatic fallback to another language or caption source. Missing tracks fail.
73
 
74
- The tool is MIT licensed. It is free to install. **FlashData API calls are paid**; installation does not include free data queries. Check current [FlashData pricing](https://flashdata.dev/pricing) and your console balance. Platform, model and customer storage charges are separate. Each search/page is a separate request; Extended Search has its own rate. Downloads are charged by delivered data volume.
75
 
76
- An uncertain timeout can still represent a charged request. Check Jobs in the FlashData console before trying again; this integration does not automatically retry paid submissions. A 401 means the key was rejected, 402 means insufficient credits, and 429 means a rate or quota limit.
77
 
78
- [Get started with this FlashData capability](https://flashdata.dev/docs/sources/youtube_transcript?utm_source=hugging-face&utm_medium=integration&utm_campaign=intent_gtm_2026q4&utm_content=youtube-transcript-research-flashdata)
 
1
  ---
2
+ title: "YouTube Transcript"
3
+ short_description: "Existing YouTube transcript text with timestamps."
4
+ emoji: "🔎"
5
+ colorFrom: "blue"
6
+ colorTo: "indigo"
7
+ sdk: "static"
8
+ app_file: "index.html"
9
  pinned: false
10
+ license: "mit"
11
  tags:
12
  - smolagents
13
  - tool
 
15
  - youtube-transcript
16
  ---
17
 
18
+ # YouTube Transcript
19
 
20
+ Retrieve existing YouTube transcript text and timestamped segments by video ID and language. Automatic captions are selected by default.
21
 
22
+ [API documentation](https://flashdata.dev/docs/sources/youtube_transcript?utm_source=hugging-face&utm_medium=integration&utm_campaign=intent_gtm_2026q4&utm_content=youtube-transcript-research-flashdata) · [Get an API key](https://flashdata.dev/app/api-keys) · [Pricing](https://flashdata.dev/pricing)
23
 
24
+ Requires your own FlashData API key and paid Credits. The Python tool is free to install.
25
 
26
+ ## Try the inputs
27
 
28
+ Edit the JSON in this Space, generate a Python call, and inspect the included sample response or paste your own JSON. This page runs locally in your browser; it does not run a data query or collect your API key.
29
 
30
+ ## Run with Python
31
+
32
+ Use Python 3.10 or newer:
33
 
34
  ```sh
35
  python -m pip install "smolagents==1.26.0" "langchain-flashdata==0.2.1"
36
  ```
37
 
38
+ Review [tool.py](tool.py) before enabling `trust_remote_code`. Run locally with your key:
39
 
40
  ```python
41
  import getpass
 
48
  print(result)
49
  ```
50
 
51
+ The same loaded tool can be used with smolagents. A direct Python call does not require a model.
52
 
53
  ## Inputs
54
 
55
+ | Parameter | Required / default | Details |
56
+ | --- | --- | --- |
57
+ | `video_id` | Required | 11-character YouTube video ID or a YouTube video URL. |
58
+ | `language` | 'en' | Requested caption language code. |
59
+ | `auto` | True | True requests automatic captions; false requests manually authored captions. |
 
 
 
 
 
 
60
 
61
+ See [input-schema.json](input-schema.json) for full validation rules. Optional `None` values are omitted; adapter defaults apply.
62
 
63
+ ## Output
64
 
65
+ Read `results.0.segments`. Fields: `start`, `end`, `text`. Optional fields may be absent.
66
 
67
+ [example.json](example.json) contains a recorded response and its original request, not a new result for the input shown in the Space.
68
 
69
+ ## Limits and cost
70
 
71
+ Defaults to existing English automatic captions. No speech recognition, translation or fallback to another track; unavailable tracks fail.
72
 
73
+ Each data request uses FlashData Credits. Check [current pricing](https://flashdata.dev/pricing) before running. Any model or hosting charges are separate.
74
 
75
+ For an uncertain timeout, check your FlashData Jobs before retrying; this tool does not automatically retry paid submissions.
index.html CHANGED
@@ -3,18 +3,18 @@
3
  <head>
4
  <meta charset="UTF-8">
5
  <meta name="viewport" content="width=device-width,initial-scale=1">
6
- <title>FlashData research tool</title>
7
  <link rel="stylesheet" href="style.css">
8
  </head>
9
  <body>
10
  <main>
11
- <header><a href="https://flashdata.dev" class="brand" target="_blank" rel="noopener">FLASHDATA</a><span>Research tools for AI agents</span></header>
12
- <section class="intro"><p class="eyebrow">SMOLAGENTS TOOL · PYTHON</p><h1 id="title">Loading tool…</h1><p id="scenario"></p><a id="docs" class="cta" target="_blank" rel="noopener">Get an API key and read the docs ↗</a></section>
13
  <div class="grid">
14
  <section class="panel"><h2>Build a request</h2><p id="task"></p><label for="parameters">Tool inputs (JSON)</label><textarea id="parameters" spellcheck="false"></textarea><button id="generate" type="button">Generate Python call</button><p id="input-status" role="status"></p><pre><code id="python"></code></pre><p class="hint">Run this code in your Python environment. Your API key is entered there, not on this page.</p></section>
15
- <section class="panel"><h2>Inspect the response</h2><p class="hint">This is a recorded documentation sample, not a live response to the request on the left. You can paste your own response JSON to explore its fields.</p><details><summary>Original sample request</summary><pre id="sample-request"></pre></details><label for="response">Response JSON</label><textarea id="response" spellcheck="false"></textarea><button id="inspect" type="button">Inspect results</button><p id="response-status" role="status"></p><div id="results"></div></section>
16
  </div>
17
- <section class="panel next"><h2>Use the result</h2><p id="next"></p><h3>Limits and costs</h3><p id="limit"></p><p>The tool is free to install. <strong>FlashData API calls are paid.</strong> Platform, model and customer storage costs are separate. No API or model call is made by this page. <a href="https://flashdata.dev/pricing" target="_blank" rel="noopener">Check pricing ↗</a></p><p>For an uncertain timeout, check Jobs before retrying a paid submission. Download jobs must be read with their saved ID; do not submit again while pending.</p></section>
18
  <footer><a id="source" target="_blank" rel="noopener">Review tool.py</a><a id="readme" target="_blank" rel="noopener">Installation and input reference</a><span>MIT licensed integration · FlashData API terms apply separately</span></footer>
19
  </main>
20
  <script type="module" src="app.js"></script>
 
3
  <head>
4
  <meta charset="UTF-8">
5
  <meta name="viewport" content="width=device-width,initial-scale=1">
6
+ <title>Search and video data tools</title>
7
  <link rel="stylesheet" href="style.css">
8
  </head>
9
  <body>
10
  <main>
11
+ <header><a href="https://flashdata.dev" class="brand" target="_blank" rel="noopener">FLASHDATA</a><span>Search &amp; video data</span></header>
12
+ <section class="intro"><p class="eyebrow">PYTHON TOOL · REQUEST EXAMPLES</p><h1 id="title">Loading tool…</h1><p id="scenario"></p><a id="docs" class="cta" target="_blank" rel="noopener">API documentation &amp; setup ↗</a></section>
13
  <div class="grid">
14
  <section class="panel"><h2>Build a request</h2><p id="task"></p><label for="parameters">Tool inputs (JSON)</label><textarea id="parameters" spellcheck="false"></textarea><button id="generate" type="button">Generate Python call</button><p id="input-status" role="status"></p><pre><code id="python"></code></pre><p class="hint">Run this code in your Python environment. Your API key is entered there, not on this page.</p></section>
15
+ <section class="panel"><h2>Inspect the response</h2><p class="hint">View the included sample or paste your own response JSON. Editing the request does not run a query.</p><details><summary>Original sample request</summary><pre id="sample-request"></pre></details><label for="response">Response JSON</label><textarea id="response" spellcheck="false"></textarea><button id="inspect" type="button">Inspect results</button><p id="response-status" role="status"></p><div id="results"></div></section>
16
  </div>
17
+ <section class="panel next"><h2>Use the result</h2><p id="next"></p><h3>Limits and costs</h3><p id="limit"></p><p>Requires your FlashData API key. <strong>Data calls use paid Credits.</strong> This page only prepares code and displays JSON. <a href="https://flashdata.dev/pricing" target="_blank" rel="noopener">Check pricing ↗</a></p><p>Check FlashData Jobs before retrying a request that timed out.</p></section>
18
  <footer><a id="source" target="_blank" rel="noopener">Review tool.py</a><a id="readme" target="_blank" rel="noopener">Installation and input reference</a><span>MIT licensed integration · FlashData API terms apply separately</span></footer>
19
  </main>
20
  <script type="module" src="app.js"></script>
task.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "slug": "youtube-transcript-research-flashdata",
3
- "title": "YouTube transcripts for cited research \u2014 FlashData",
4
- "scenario": "Give a research agent timestamped text from an existing YouTube caption track. Keep segment start and end times so a reader can return to the video when checking a generated summary.",
5
- "task": "Read existing English automatic captions for a timestamped video brief.",
6
  "input": {
7
  "video_id": "FUfGcZ092b0",
8
  "language": "en",
@@ -14,8 +14,9 @@
14
  "end",
15
  "text"
16
  ],
17
- "next": "Retain video ID, language and timestamps when passing segments to your own model. Model calls and summarization are separate steps.",
18
- "limit": "Defaults to existing automatic English captions. No ASR, translation or automatic fallback to another language or caption source. Missing tracks fail.",
 
19
  "source": "youtube_transcript",
20
  "repo_id": "flashdata-dev/youtube-transcript-research-flashdata"
21
  }
 
1
  {
2
  "slug": "youtube-transcript-research-flashdata",
3
+ "title": "YouTube Transcript",
4
+ "scenario": "Retrieve existing YouTube transcript text and timestamped segments by video ID and language. Automatic captions are selected by default.",
5
+ "task": "Fetch an existing transcript for a video.",
6
  "input": {
7
  "video_id": "FUfGcZ092b0",
8
  "language": "en",
 
14
  "end",
15
  "text"
16
  ],
17
+ "next": "Read results.0.segments for text with start and end timestamps.",
18
+ "limit": "Defaults to existing English automatic captions. No speech recognition, translation or fallback to another track; unavailable tracks fail.",
19
+ "short_description": "Existing YouTube transcript text with timestamps.",
20
  "source": "youtube_transcript",
21
  "repo_id": "flashdata-dev/youtube-transcript-research-flashdata"
22
  }
tool.py CHANGED
@@ -3,7 +3,7 @@ from smolagents import Tool
3
 
4
  class FlashDataTool(Tool):
5
  name = 'youtube_transcript'
6
- description = 'Give a research agent timestamped text from an existing YouTube caption track. Keep segment start and end times so a reader can return to the video when checking a generated summary. Defaults to existing automatic English captions. No ASR, translation or automatic fallback to another language or caption source. Missing tracks fail. Uses paid FlashData API credits. Never automatically retry an uncertain submission; check Jobs first.'
7
  inputs = {'auto': {'description': 'True requests automatic captions; false requests manually authored '
8
  'captions. Default: True.',
9
  'nullable': True,
 
3
 
4
  class FlashDataTool(Tool):
5
  name = 'youtube_transcript'
6
+ description = 'Retrieve existing YouTube transcript text and timestamped segments by video ID and language. Automatic captions are selected by default. Defaults to existing English automatic captions. No speech recognition, translation or fallback to another track; unavailable tracks fail. Uses paid FlashData API credits. Never automatically retry an uncertain submission; check Jobs first.'
7
  inputs = {'auto': {'description': 'True requests automatic captions; false requests manually authored '
8
  'captions. Default: True.',
9
  'nullable': True,