docs: Add NeatLogs integration videos to documentation #5940
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,134 +1,132 @@ | ||
| --- | ||
| title: Neatlogs Integration | ||
| description: Collaborative debugging and observability for your CrewAI agents | ||
| icon: magnifying-glass-chart | ||
| mode: "wide" | ||
| --- | ||
|
|
||
| [Neatlogs](https://neatlogs.com/?utm_source=crewAI-docs) is a collaborative debugging workspace for AI agent teams. It captures every LLM call, tool invocation, token count, and latency and surfaces it all in a shared dashboard your whole team can search, comment on, and act from. | ||
|
|
||
| <iframe | ||
| className="w-full aspect-video rounded-xl" | ||
| src="https://www.youtube.com/embed/dfVGAp7xVC8" | ||
| title="Neatlogs overview" | ||
| frameBorder="0" | ||
| allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" | ||
| allowFullScreen | ||
| /> | ||
|
|
||
| ## What is Neatlogs? | ||
|
|
||
| {/* ADD YOUTUBE VIDEO HERE - team to provide URL */} | ||
| {/* | ||
| <iframe | ||
| className="w-full aspect-video rounded-xl" | ||
| src="https://www.youtube.com/embed/YOUR_VIDEO_ID" | ||
| title="What is Neatlogs" | ||
| frameBorder="0" | ||
| allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" | ||
| allowFullScreen | ||
| /> | ||
| */} | ||
|
|
||
| Agent failures rarely throw exceptions. They produce wrong outputs, miss tool calls, or hallucinate. Neatlogs gives your team the full picture: what the agent decided, why, and where it went wrong. | ||
|
|
||
| <CardGroup cols={2}> | ||
| <Card title="Trace Viewer" icon="list-tree"> | ||
| Follow every thought, tool call, and decision in sequence. Token counts and latency included. | ||
| </Card> | ||
| <Card title="AI-Powered Search" icon="sparkles"> | ||
| Find traces, runs, and patterns in plain English. No query language needed. | ||
| </Card> | ||
| <Card title="Inline Comments" icon="message"> | ||
| Tag teammates on any trace step. Threads keep investigation in one place. | ||
| </Card> | ||
| <Card title="Feedback & Evaluation" icon="thumbs-up"> | ||
| Mark outputs correct or incorrect. Build eval datasets straight from production traces. | ||
| </Card> | ||
| <Card title="Ask the Trace" icon="robot"> | ||
| Chat with your trace using the built-in AI assistant to surface root causes fast. | ||
| </Card> | ||
| <Card title="Public Sharing" icon="share-nodes"> | ||
| Publish trace links to your community or stakeholders for async review. | ||
| </Card> | ||
| </CardGroup> | ||
|
|
||
| ## Quick Setup | ||
|
|
||
| <Steps> | ||
| <Step title="Sign up and get your API key"> | ||
| Go to [neatlogs.com](https://neatlogs.com/?utm_source=crewAI-docs), create a project, and copy the API key from your dashboard. | ||
| </Step> | ||
|
Comment on lines
+59
to
+61
There was a problem hiding this comment. 🧩 Analysis chain🌐 Web query:
💡 Result: The neatlogs.init function in the Neatlogs Python SDK is used to initialize the SDK and configure tracking for your application [1][2]. It should be called at the very beginning of your entrypoint script, before importing any instrumented libraries (such as LangChain or OpenAI), to ensure the SDK can properly patch and trace those libraries [3][1]. Required and Optional Parameters: The primary required parameter is: - api_key (str): Your Neatlogs API key [1][2]. Common optional parameters include: - endpoint (str): A custom API endpoint URL. If you need to override the default, the typical value used in documentation is https://api.neatlogs.com/v4/batch [1]. - instrumentations (list[str]): A list of frameworks to automatically instrument (e.g., ["openai", "langchain", "crewai"]) [3][1]. - workflow_name (str): A name for the workflow or project [3][1]. - metadata (dict): A dictionary for global metadata (e.g., {"env": "production"}) [1]. - tags (list[str]): A list of tags to associate with the traces [2]. - auto_session (bool): Set to True to enable automatic session grouping for traces [4]. Regarding the endpoint value: For most standard use cases, you do not need to provide an endpoint parameter at all; the SDK defaults to the official Neatlogs platform [1]. If you are required to use a specific or custom endpoint (such as for on-premise installations or proxying), this value is provided by your Neatlogs account administrator or via your organization's internal documentation [1][5]. The default value often cited in SDK examples is https://api.neatlogs.com/v4/batch [1]. Citations:
🏁 Script executed: #!/bin/bash
set -euo pipefail
FILE="docs/en/observability/neatlogs.mdx"
echo "== File: $FILE =="
wc -l "$FILE"
# Print lines around 45-90 to capture Step 1 and the init snippet
nl -ba "$FILE" | sed -n '40,110p'Repository: crewAIInc/crewAI Length of output: 207 🏁 Script executed: #!/bin/bash
set -euo pipefail
FILE="docs/en/observability/neatlogs.mdx"
echo "== Lines 45-90 =="
cat -n "$FILE" | sed -n '45,90p'
echo
echo "== Lines 90-130 =="
cat -n "$FILE" | sed -n '90,130p'Repository: crewAIInc/crewAI Length of output: 4438 🏁 Script executed: #!/bin/bash
set -euo pipefail
rg -n "NEATLOGS_ENDPOINT" -S . || true
rg -n "endpoint\s*=" -S docs/en/observability/neatlogs.mdx || true
rg -n "NEATLOGS_API_KEY|NEATLOGS_ENDPOINT" -S docs/en/observability/neatlogs.mdx || trueRepository: crewAIInc/crewAI Length of output: 307 Document “Sign up and get your API key” (lines 59-61) only tells users to copy 🤖 Prompt for AI Agents |
||
| <Step title="Install the SDK"> | ||
| ```bash | ||
| pip install neatlogs | ||
| ``` | ||
| </Step> | ||
| <Step title="Add this before your Crew runs"> | ||
| Call `neatlogs.init()` before importing CrewAI so the auto-instrumentation patches land first. | ||
|
|
||
| ```python | ||
| import os | ||
| import neatlogs | ||
|
|
||
| neatlogs.init( | ||
| api_key=os.environ["NEATLOGS_API_KEY"], | ||
| endpoint=os.environ["NEATLOGS_ENDPOINT"], | ||
| workflow_name="my-agent", | ||
| instrumentations=["crewai"], | ||
| ) | ||
|
|
||
| # Now import and run your crew as normal | ||
| from crewai import Crew, Agent, Task | ||
| ``` | ||
|
|
||
| That's it — no changes to the rest of your code. | ||
| </Step> | ||
| <Step title="Traces logged directly in your dashboard"> | ||
| Every tool call, agent run, and LLM call is captured and surfaced in your Neatlogs dashboard automatically. Tag teammates and leave comments directly on any trace step — no context switching, no Slack threads, no copy-pasting outputs. | ||
|
|
||
| <video autoPlay loop muted playsInline className="w-full rounded-xl"> | ||
| <source src="/images/neatlogs-traces.mp4" type="video/mp4" /> | ||
| </video> | ||
|
Comment on lines
+90
to
+92
There was a problem hiding this comment. 🧩 Analysis chain🏁 Script executed: #!/bin/bash
# Verify that all referenced video files exist in the repository
echo "Checking for video files..."
for video in neatlogs-traces.mp4 neatlogs-experiments.mp4 neatlogs-codefixes.mp4 neatlogs-evals.mp4; do
if fd -t f "^${video}$" docs/images/; then
echo "✓ Found: ${video}"
else
echo "✗ Missing: ${video}"
fi
doneRepository: crewAIInc/crewAI Length of output: 314 🏁 Script executed: #!/bin/bash
set -euo pipefail
echo "Checking exact file paths:"
for video in neatlogs-traces.mp4 neatlogs-experiments.mp4 neatlogs-codefixes.mp4 neatlogs-evals.mp4; do
p="docs/images/$video"
if [ -f "$p" ]; then
echo "✓ Present: $p"
else
echo "✗ Missing: $p"
fi
doneRepository: crewAIInc/crewAI Length of output: 266 Fix missing video asset for neatlogs-traces.
🤖 Prompt for AI Agents |
||
| </Step> | ||
| <Step title="Experiments — version control for your prompts"> | ||
| Iterate on prompts without touching production. Neatlogs gives you version control for every prompt in your crew and a playground to test changes side by side — so you can improve agent behaviour safely before it goes anywhere near your live system. | ||
|
|
||
| <video autoPlay loop muted playsInline className="w-full rounded-xl"> | ||
| <source src="/images/neatlogs-experiments.mp4" type="video/mp4" /> | ||
| </video> | ||
| </Step> | ||
| <Step title="Code fixes directly in your IDE"> | ||
| Get agent improvement suggestions and code fixes surfaced straight into Cursor, Codex, Claude Code, and other AI-native editors. Debug and improve your crew without leaving the tools you already work in. | ||
|
|
||
| <video autoPlay loop muted playsInline className="w-full rounded-xl"> | ||
| <source src="/images/neatlogs-codefixes.mp4" type="video/mp4" /> | ||
| </video> | ||
| </Step> | ||
| <Step title="Human evaluations"> | ||
| Bring your team in to review agent outputs, mark runs as correct or incorrect, and build evaluation datasets straight from production traces. Close the loop between what your agents produce and what they should produce. | ||
|
|
||
| <video autoPlay loop muted playsInline className="w-full rounded-xl"> | ||
| <source src="/images/neatlogs-evals.mp4" type="video/mp4" /> | ||
| </video> | ||
| </Step> | ||
| </Steps> | ||
|
|
||
| ## Resources | ||
|
|
||
| <CardGroup cols={2}> | ||
| <Card title="Neatlogs Docs" icon="book" href="https://docs.neatlogs.com/"> | ||
| Full SDK reference and integration guides | ||
| </Card> | ||
| <Card title="Dashboard & API Key" icon="key" href="https://neatlogs.com/?utm_source=crewAI-docs"> | ||
| Create a project and get your API key | ||
| </Card> | ||
| <Card title="GitHub SDK" icon="github" href="https://github.com/NeatLogs/neatlogs"> | ||
| MIT licensed · Python 3.8+ | ||
| </Card> | ||
| <Card title="Contact" icon="envelope" href="mailto:hello@neatlogs.com"> | ||
| hello@neatlogs.com | ||
| </Card> | ||
| </CardGroup> | ||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Remove commented-out TODO code from production documentation.
The commented placeholder and TODO note should be removed. A YouTube video is already embedded in the hero section above (lines 10-17), making this TODO obsolete and confusing.
🧹 Proposed cleanup
📝 Committable suggestion
🤖 Prompt for AI Agents