Streaming Telemetry & HLS Diagnostics MCP
Diagnose HLS playlists (.m3u8), audit SCTE-35 ad breaks, and triage playback QoE.
Free
Works with the AI Agents you already use
Streaming Telemetry & HLS Diagnostics MCP
Example session with this skill installed
Inspect this HLS master manifest and tell me if our bitrate ladder has any aggressive step jumps that could cause player rebuffering stalls.
Analyze this HLS manifest at https://cdn.example.com/live/stream.m3u8 and check for SCTE-35 ad marker synchronization issues reported in recent session drops.
- Read your context and instructions
- Compiled the streaming telemetry
Summary
Analyzed Master Playlist containing 3 renditions (360p, 720p, 1080p). Identified 2 high-risk bitrate ladder steps exceeding the recommended 2.5x threshold.
| Rendition | Resolution | Bandwidth (bps) | Step Ratio vs Previous | Compliance Status |
|---|---|---|---|---|
| 360p | 640x360 | 800,000 | Baseline | ✅ Optimal |
| 720p | 1280x720 | 2,400,000 | 3.0x jump | ⚠️ High Risk |
| 1080p | 1920x1080 | 7,500,000 | 3.1x jump | ⚠️ High Risk |
Detected Video Playback Risks:
Aggressive Bitrate Step (3.0x): Jump from 800 kbps (360p) to 2.4 Mbps (720p) exceeds the 2.5x threshold, risking player buffer depletion during sudden network throttling.
Aggressive Bitrate Step (3.1x): Jump from 2.4 Mbps (720p) to 7.5 Mbps (1080p) risks visual stutter and rebuffering stalls on mobile clients.
Recommended Next Steps:
- Insert an intermediate 480p rendition (approx. 1.4 Mbps) to smooth the lower ladder step.
- Insert an intermediate 1080p medium-profile rendition (approx. 4.5 Mbps) before stepping up to 7.5 Mbps.
Connects securely to your tools. The creator never sees your data.
About this skill
Equips Claude Code, Cursor, Copilot, and other AI coding agents with direct video streaming diagnostics to inspect HLS playlists, validate SCTE-35 ad break splices, and triage playback QoE incidents.
Backed by the open-source streaming-telemetry-mcp server, this skill allows your AI assistant to run local diagnostics without sending proprietary stream URLs or internal telemetry to external third-party cloud APIs.
What Your Agent Can Do With This Skill:
- HLS Playlist Diagnostics: Ingests Master and Media
.m3u8manifests to audit bitrate ladder steps (flagging aggressive bandwidth drops > 2.5x that cause player ABR buffering stalls), check#EXT-X-TARGETDURATIONsegment overruns, and detect discontinuity tags. - SCTE-35 & SSAI Ad Insertion Validation: Audits
#EXT-X-DATERANGEand#EXT-X-CUE-OUT/#EXT-X-CUE-INsplice synchronization to detect unclosed ad breaks and planned duration mismatches before broadcast. - Playback QoE Incident Triage: Evaluates playback telemetry (startup time/TTFF, rebuffering ratio %, CDN edge POP, error codes like CDN 504 or HTTP 503) and generates an automated Root Cause Analysis (RCA) with mitigation and multi-CDN traffic steering recommendations.
Setup & Local Execution:
Includes client configuration templates for Cursor (.cursor/mcp.json) and Claude Desktop (claude_desktop_config.json). The MCP server runs 100% locally on your machine.
- Upstream Open-Source Repository: https://github.com/soniran635/streaming-telemetry-mcp
- License: MIT License (Free & Open Source)
How it works
- Analyze Manifest via the
inspect_hls_manifesttool to scan for spec compliance and encoding issues. - Validate Ad Markers using
validate_ad_markersto ensure ad pods are correctly signaled and closed. - Query Telemetry with
query_stream_qoeto correlate manifest issues with real-world player errors like CDN_504.
Frameworks & tools
This skill requires the streaming-telemetry-mcp server. It supports HLS (.m3u8) video standards, SCTE-35 ad-insertion protocols, and standard QoE telemetry metrics including TTFF and rebuffer ratios.
Why this beats prompting it yourself
Generic LLMs cannot parse multi-megabyte HLS playlists or perform precise duration arithmetic for ad-splice validation without hallucinating timing offsets. This skill connects the agent to specialized tools that extract exact manifest data and telemetry trends for objective debugging.
Use cases
- Debugging playback stalls caused by EXT-X-DISCONTINUITY tags.
- Verifying SCTE-35 splice points for a new SSAI integration.
- Performing automated RCA on player telemetry spikes during live events.
- Auditing bitrate ladders for spec compliance before deployment.
Known limitations
Requires a local instance of the streaming-telemetry-mcp server configured in your agent settings. Does not support DASH (.mpd) manifests or DRM license challenge inspection.
Requires a local streaming-telemetry-mcp server instance. Limited to HLS (.m3u8) and SCTE-35; does not support DASH or DRM-encrypted header analysis.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account
Permissions
No special permissions declared or detected