ttagent
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| extract_videoA | Harvest a public TikTok video: given a video/photo URL, short link or bare video id, decodes the public surface and downloads the video (HD when exposed), the soundtrack and covers (all magic-byte verified) to out_dir. Returns the summary plus the full video_manifest envelope (author, title, stats, music identity, media URLs, local files, errors, decode-slot provenance). No login, no API keys. |
| lookup_videoA | Fast metadata-only lookup of a public TikTok video: same decode pipeline as extract_video but with --no-download — author, title, stats, music identity, media URLs, without writing media files. Use this when you only need to read the video. |
| read_manifestA | Read an existing video_manifest.json produced by a previous extract/lookup and return the full envelope. Refuses any file not named video_manifest.json. |
| get_schemaA | Return the JSON Schema (draft-07) describing the video_manifest envelope (schema_version 1.0) — use it to validate or explore the output contract. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: extract_video downloads media, lookup_video fetches metadata only, read_manifest reads a stored manifest, and get_schema returns the contract schema. The extract/lookup pair is well-differentiated by the download flag and use-case wording, leaving no realistic ambiguity.
All four tools follow a clean verb_noun pattern: extract_video, lookup_video, read_manifest, get_schema. No mixed conventions or vague verbs.
Four tools is tight and well-scoped for a TikTok public video harvester: one for full extraction, one for metadata, one for reading stored output, and one for schema introspection. Every tool earns its place.
The core lifecycle for a single public video — fetch metadata, extract media, persist and re-read a manifest, and inspect the output contract — is covered. Missing only peripheral operations like listing or discovery, which are outside the stated per-URL harvesting scope.