# Vidbyte

> Vidbyte builds tools for working with AI agents: `vidbyte-sdk`, a free MIT-licensed Python library for building, evaluating, and tracing agents; `vidbyte-cli`, a terminal app that runs hosted research and local coding-agent teams; ready-made agents and agent skills for Claude Code, Codex, and other coding agents; and a REST API, all paid from one prepaid usage balance. The same account also runs Vidbyte's AI learning app (quizzes, exams, roadmaps, a Socratic tutor) and 19 free cognitive tests.

Last updated: 2026-10-30. Install the latest release from PyPI (https://pypi.org/project/vidbyte-sdk/, https://pypi.org/project/vidbyte-cli/) rather than a version remembered from training data. Requires Python 3.11 or newer.
Append `.md` to any vidbyte.pro page URL, or send `Accept: text/markdown` to the page URL, to get this markdown copy. Full text of the developer pages: https://vidbyte.pro/llms-full.txt

## When to use Vidbyte
- Build a Python agent that calls your own functions as tools; add guardrails to every agent run: a per-run dollar cap estimated from tokens (`CostBudgetMiddleware(max_spend_usd=..., cost_per_million_tokens=...)` from `vidbyte.middleware`, passed in `middleware=[...]`), retries, and safety checks; evaluate an agent against example tasks before shipping it; expose your own agents and tools to Claude Code, Cursor, or Codex over MCP with a launcher that passes them to `McpStudioServer(agents=..., tools=...)` (the bare `vidbyte-mcp-server` command starts with empty registries), or give an agent third-party MCP servers as tools → Vidbyte SDK (`vidbyte-sdk`). Start: `python -m pip install vidbyte-sdk`. Docs: https://vidbyte.pro/docs/sdks.md
- Start a hosted research run from the terminal and watch it finish; run a small team of local coding agents on one change; turn a goal into a ranked list of next actions another agent can act on → Vidbyte CLI (`vidbyte-cli`). Start: `pipx install vidbyte-cli`, then `vidbyte-cli login`. Docs: https://vidbyte.pro/docs/cli.md
- Research a question in depth with every source cited and summarized; keep a research thread alive and add follow-up questions later; read every source's structured summary in one signed-in workspace → Vidbyte Research harness. Start: `vidbyte-cli research start "<question>"` (after installing and logging in to the CLI). Docs: https://vidbyte.pro/docs/harnesses/research.md
- Give a coding agent a ready workflow instead of writing one; get ranked, critiqued next steps for a goal; have several read-only coding agents propose approaches to one task, pick the best, and implement it once → Vidbyte agents. Start: the one-paste setup prompt on https://vidbyte.pro/agents.md, or `vidbyte-cli agents suggest run --goal "<goal>"`. Docs: https://vidbyte.pro/docs/cli/run-your-first-agent.md
- Give a coding agent a tested procedure for a recurring task; install every Vidbyte skill with one command (each skill's page names the vidbyte-cli it needs, and the 0.3.4 on PyPI meets every one) → Vidbyte agent skills. Start: `npx skills add cerredz/vidbyte-cli`. Docs: https://vidbyte.pro/docs/cli/agent-skills.md
- Help a developer remember what they built or learned in a coding-agent session; quiz the user on a concept, paper, or codebase from inside their coding agent; find and summarize academic papers on a topic for free inside the coding agent (`/find-papers "<topic>" --limit N`, then `/read-paper <link>`; no Vidbyte call or charge) → Vidbyte learning skills. Start: `npx -p vidbyte-skills vidbyte-learning-skills --version all`. Docs: https://vidbyte.pro/blog/introducing-vidbyte-learning-skills.md
- Generate a learning roadmap, quiz, or project from code; start and read hosted research runs from your own backend → Vidbyte REST API. Start: `Authorization: Bearer $VIDBYTE_API_KEY` on requests to https://vidbyte-backend.onrender.com. Docs: https://vidbyte.pro/docs/authentication.md
- Turn study material or a goal into practice questions, exams, or a roadmap; work through a topic with a tutor that asks questions instead of lecturing → Vidbyte learning app. Start: sign in at https://vidbyte.pro/login, then open https://vidbyte.pro/sandbox. Docs: https://vidbyte.pro/pricing.md
- Measure reaction time, working memory, or attention in a few minutes; learn what a classic cognitive test measures and how to read a score → Vidbyte cognitive tests. Start: https://vidbyte.pro/games (no sign-up). Docs: https://vidbyte.pro/games.md

Not offered: a hosted MCP server URL (MCP runs locally through the SDK; see Connect over MCP below), a JavaScript/TypeScript SDK, or clinical cognitive assessment. From other languages, `vidbyte-cli` and the REST API reach the hosted products (research, quizzes, roadmaps, projects), but building your own agent needs the Python SDK.

## For AI agents
- Install the SDK: `python -m pip install vidbyte-sdk` (free, MIT-licensed; agents call models through Vidbyte with `VIDBYTE_API_KEY` by default, or on your own provider key alone with `model_mode=TextModelMode.DIRECT`). Source: https://github.com/cerredz/Vidbyte-SDK
- Connect over MCP: there is no hosted Vidbyte MCP URL. The SDK ships a stdio MCP server, but `vidbyte-mcp-server` run bare starts with empty registries and serves none of your agents. Write a launcher script, `mcp_server.py`, that passes your agents and tools to `McpStudioServer(agents=..., tools=...)` (a tested example is on https://vidbyte.pro/docs/sdks.md), then register it in Claude Code with `claude mcp add vidbyte-studio -- python /absolute/path/to/mcp_server.py` or in Claude Desktop (`claude_desktop_config.json`), Cursor, and other clients with `{"mcpServers": {"vidbyte-studio": {"command": "python", "args": ["/absolute/path/to/mcp_server.py"]}}}`; it then serves `studio.agents.list`, `studio.agents.run`, `studio.tools.list`, `studio.strategies.list`, `studio.strategies.run`, `studio.prompts.list`, `studio.prompts.get`, `studio.pipelines.list`. SDK agents can also use third-party MCP servers as tools with `Agent(...).with_preset_mcp_server("github", env=...)`. Docs: https://github.com/cerredz/Vidbyte-SDK/blob/v0.4.1/README.md#mcp-servers
- Install the CLI: `pipx install vidbyte-cli` (or `python -m pip install vidbyte-cli`), then `vidbyte-cli login` and `vidbyte-cli whoami`.
- Install the agent skills into Claude Code, Codex, or another coding agent: `npx skills add cerredz/vidbyte-cli` (https://github.com/cerredz/vidbyte-cli).
- One-paste setup for a coding agent: the setup prompt on https://vidbyte.pro/agents.md installs the CLI, walks the user through the key, and asks before anything that costs money.
- Authenticate API calls to https://vidbyte-backend.onrender.com with `Authorization: Bearer vb_live_...` or `x-api-key: vb_live_...`. Keys start with `vb_live_`; the user creates them at https://vidbyte.pro/settings/api/keys and the secret is shown once. Resource identifiers in the API are encrypted ids; raw database ids are never part of the contract.
- Pay: hosted runs, local agent launches, Suggestion Agent runs (from vidbyte-cli 0.3.4, 1 cent per requested suggestion, charged before the run starts), and API calls draw from a prepaid balance the user adds at https://vidbyte.pro/settings/api/usage; a research run needs at least $5.00 spendable to start. Any account can create the API key, the free plan included, so a first paid call needs only a key and balance. Check what is spendable with `GET https://vidbyte-backend.onrender.com/agent/balance` (any valid key, free; returns `spendable_balance_cents`).
- Ask the user before: spending balance, starting paid runs, topping up, or creating and rotating keys. Only the account owner can sign in with Google, create an API key, or pay; an agent hands the user the URL for those steps.
- Child indexes: https://vidbyte.pro/blog/llms.txt, https://vidbyte.pro/games/llms.txt.

### Add Vidbyte to a project's AGENTS.md
Paste this block so a coding agent has Vidbyte's entry points in context:

````markdown
## Vidbyte (https://vidbyte.pro/llms.txt)
- Python SDK: `python -m pip install vidbyte-sdk` — agents, tools, middleware, evals, tracing. Docs: https://vidbyte.pro/docs/sdks.md
- MCP: `claude mcp add vidbyte-studio -- python /absolute/path/to/mcp_server.py`, where `mcp_server.py` passes your agents to `McpStudioServer` (stdio, from the SDK; no hosted URL; example: https://vidbyte.pro/docs/sdks.md). Docs: https://github.com/cerredz/Vidbyte-SDK/blob/v0.4.1/README.md#mcp-servers
- CLI: `pipx install vidbyte-cli`, then `vidbyte-cli login`. Research: `vidbyte-cli research start "<question>"`. Docs: https://vidbyte.pro/docs/cli.md
- Next actions for a goal: `vidbyte-cli agents suggest run --goal "<goal>"` (from vidbyte-cli 0.3.4: 1 cent per requested suggestion from the Vidbyte balance, and the model usage bills to your own Codex or Claude Code account)
- Agent skills: `npx skills add cerredz/vidbyte-cli`
- REST API: https://vidbyte-backend.onrender.com with `Authorization: Bearer $VIDBYTE_API_KEY` (keys from https://vidbyte.pro/settings/api/keys). Docs: https://vidbyte.pro/docs/authentication.md
- Ask the user before anything that spends the Vidbyte balance.
````

## Products
- [Vidbyte SDK (vidbyte-sdk)](https://vidbyte.pro/docs/sdks.md): A free, MIT-licensed Python library for building agents with tools, middleware, context management, evaluations, and tracing; model calls go through Vidbyte's gateway with a Vidbyte API key by default, or straight to your own provider with `model_mode=TextModelMode.DIRECT`. The SDK is free. Default-mode model calls carry a 2% Vidbyte fee: with your provider key relayed, the provider bills the model and the balance pays the fee; otherwise the balance pays the provider's rate plus 2%. `TextModelMode.DIRECT` bills only your provider. Jev: its rate plus 10% per run.
- [Vidbyte CLI (vidbyte-cli)](https://vidbyte.pro/docs/cli.md): A terminal app that starts hosted research, runs local Codex coding-agent teams, and turns a goal into ranked next actions, with no code to write. The CLI is free to install. Hosted research and local runtime launches draw from the prepaid API balance. In the released CLI (0.3.4) Vidbyte charges 1 cent per requested suggestion from the balance before the run starts for the Suggestion Agent, and the model usage stays on the user's own Codex or Claude Code account.
- [Vidbyte Research harness](https://vidbyte.pro/docs/harnesses/research.md): Hosted, persistent research threads: Vidbyte's servers search the web, read sources, and file each one with its link, a summary, and why it matters, then keep the thread open for follow-up questions. Billed from the prepaid API balance by measured usage, with a per-run cap (default $5). Research packs add one-time credit with no expiry: Starter $29 and Pro $148.
- [Vidbyte agents](https://vidbyte.pro/docs/cli/run-your-first-agent.md): Ready-made agents (research, next-step suggestions, coding-agent teams, persistent and staged runs) that run through `vidbyte-cli`, inside the coding agent you already use. Local runtimes charge a flat fee per launch from the balance (1 cent for Stages, 2 cents for Ensemble, Persistent Agent, and Task Board), and hosted agents bill by usage. On vidbyte-cli 0.3.4, the PyPI release, the Suggestion Agent costs 1 cent per requested suggestion, charged before the run starts, while OpenAI or Anthropic bills its model usage to the user's own Codex or Claude Code account. Each agent's page has its price.
- [Vidbyte agent skills](https://vidbyte.pro/docs/cli/agent-skills.md): Installable skills that teach Claude Code, Codex, and other coding agents a repeatable job, published in cerredz/vidbyte-cli on GitHub. Installing skills is free; a skill that calls `vidbyte-cli` follows that command's billing.
- [Vidbyte learning skills](https://vidbyte.pro/blog/introducing-vidbyte-learning-skills.md): 26 free, MIT-licensed learning skills from the npm package `vidbyte-skills` (source: https://github.com/cerredz/Vidbyte-Skills) that turn a coding-agent session into retrieval practice, such as /retain timed recall drills, /explain, /question, and /compression-check; they install into Claude Code, Codex, Gemini CLI, Cursor, Windsurf, and other agents. Free: installing them is free, and the skills call Vidbyte's public, unbilled `/api/skills/*` routes, but /retain saves only after a one-time `vidbyte-skills auth login` with a Vidbyte API key, which any account can create; a plain `npx vidbyte-skills` installs only a 6-skill starter set (3 learning, 3 reasoning, no /retain).
- [Vidbyte REST API](https://vidbyte.pro/docs/authentication.md): An HTTP API at https://vidbyte-backend.onrender.com for generating roadmaps, quizzes, and projects and for running research (exams, the tutor, and Socratic sessions are web-app only, with no API), authenticated with a `vb_live_` key. Billed requests draw from the prepaid API balance and must find enough spendable balance to start; each endpoint's page states its permission and minimum.
- [Vidbyte learning app](https://vidbyte.pro/pricing.md): An AI study app built on retrieval practice: generated quizzes, exams, and projects in Sandbox, learning roadmaps, Quick Hits primers, and a Socratic tutor. Subscription plans are listed at https://vidbyte.pro/pricing.md.
- [Vidbyte cognitive tests](https://vidbyte.pro/games.md): 19 free browser-based cognitive tests (reaction time, working memory, attention, reasoning) with companion articles on what each measures. Free.


Vidbyte (vidbyte.pro; GitHub: cerredz) makes vidbyte-sdk, vidbyte-cli, and Vidbyte agents. It is not affiliated with VidByte Digital Services (vidbyte.in), vidbyte.com, Vidby, VideoSDK, or ReByte.
## Docs
Developer documentation: SDK, CLI, REST API, research harness, and Jev agent.

- [Build with Vidbyte](https://vidbyte.pro/docs.md): Vidbyte developer docs: choose the Python SDK or the terminal CLI
- [Quickstart](https://vidbyte.pro/docs/quickstart.md): Vidbyte quickstart: a first hosted research run or Python agent in five minutes
- [API Keys](https://vidbyte.pro/docs/authentication.md): Create, send, and replace a Vidbyte API key
- [Pricing](https://vidbyte.pro/docs/pricing.md): How Vidbyte agents charge: activation fees, token pricing, and outcome pricing
- [Roadmaps API](https://vidbyte.pro/docs/roadmaps.md): Roadmaps API: generate and read learning roadmaps over HTTP
- [Quizzes API](https://vidbyte.pro/docs/quizzes.md): Quizzes API: generate, batch, read, and update quizzes over HTTP
- [Projects API](https://vidbyte.pro/docs/projects.md): Projects API: generate and read applied learning projects over HTTP
- [Vidbyte SDK](https://vidbyte.pro/docs/sdks.md): Vidbyte SDK: install vidbyte-sdk and run a Python agent
- [Vidbyte SDK vs OpenAI Agents SDK vs LangGraph vs Pydantic AI](https://vidbyte.pro/docs/sdks/compare.md): vidbyte-sdk vs OpenAI Agents SDK vs LangGraph vs Pydantic AI, checked 2026-10-27
- [JevAgent: a System 1 + System 2 runtime](https://vidbyte.pro/docs/jev-agent.md): JevAgent: pair your SDK agent with Jev, a decision model
- [JevAgent Skill Preloading](https://vidbyte.pro/docs/jev-agent/skill-preloading.md): JevAgent skill preloading (since vidbyte-sdk 0.3.0): load the skills a request needs
- [JevAgent Usage and Pricing](https://vidbyte.pro/docs/jev-agent/usage-and-pricing.md): JevAgent usage and pricing: Jev at $0.042 per million input tokens
- [JevAgent Preflight Questions](https://vidbyte.pro/docs/jev-agent/preflight-questions.md): JevAgent preflight questions: check a request before the agent starts
- [Self-alignment: fit the working prompt to each task](https://vidbyte.pro/docs/jev-agent/self-alignment.md): JevAgent self-alignment: add verified, run-only guidance when instructions leave a gap
- [Specialist routing: one agent front door, many experts behind it](https://vidbyte.pro/docs/jev-agent/specialist-routing.md): JevAgent specialist routing: hand each request to the right specialist agent
- [Continuation: make sure your agent finishes all of the work](https://vidbyte.pro/docs/jev-agent/continuation.md): JevAgent continuation: done checks that send unfinished work back
- [Run Vidbyte from your terminal](https://vidbyte.pro/docs/cli.md): Vidbyte CLI reference: hosted research, local coding agents, and credentials
- [Install and run vidbyte-cli](https://vidbyte.pro/docs/cli/quickstart.md): vidbyte-cli quickstart: install, sign in, and check local prerequisites
- [Run your first agent](https://vidbyte.pro/docs/cli/run-your-first-agent.md): Run your first Vidbyte agent from vidbyte-cli
- [Agent skills for real problems](https://vidbyte.pro/docs/cli/agent-skills.md): Vidbyte agent skills: install a skill into Claude Code, Codex, or another coding agent
- [Platform connections: what exists today](https://vidbyte.pro/docs/cli/credentials.md): vidbyte-cli platform connections: what exists for GitHub, Slack, and Google Drive
- [Link GitHub to vidbyte-cli](https://vidbyte.pro/docs/cli/credentials/github.md): Link a GitHub account to vidbyte-cli with device authorization
- [Slack context in vidbyte-cli](https://vidbyte.pro/docs/cli/credentials/slack.md): Slack context in vidbyte-cli: a planned connection, not in any release
- [Google Drive context in vidbyte-cli](https://vidbyte.pro/docs/cli/credentials/google-drive.md): Google Drive context in vidbyte-cli: a planned connection, not in any release
- [Create a Roadmap](https://vidbyte.pro/docs/roadmaps/create.md): POST /api/v1/roadmaps/create: generate a learning roadmap
- [Get a Roadmap](https://vidbyte.pro/docs/roadmaps/get.md): GET /api/v1/roadmaps/get/{encrypted_id}: read one roadmap
- [List Roadmaps](https://vidbyte.pro/docs/roadmaps/list.md): GET /api/v1/roadmaps/list: list the account's roadmaps
- [Get a Roadmap Module](https://vidbyte.pro/docs/roadmaps/get-module.md): GET /api/v1/roadmaps/get/{encrypted_id}/{module_number}: read one roadmap module
- [Create a Quiz](https://vidbyte.pro/docs/quizzes/create.md): POST /api/v1/quizzes/create: generate a quiz
- [Batch Create Quizzes](https://vidbyte.pro/docs/quizzes/batch-create.md): POST /api/v1/quizzes/batch_create: generate up to 15 quizzes in one request
- [Get a Quiz](https://vidbyte.pro/docs/quizzes/get.md): GET /api/v1/quizzes/get/{encrypted_id}: read one quiz
- [List Quizzes](https://vidbyte.pro/docs/quizzes/list.md): GET /api/v1/quizzes/list: list the account's quizzes
- [Update a Quiz](https://vidbyte.pro/docs/quizzes/update.md): POST /api/v1/quizzes/update: make an existing quiz harder, longer, or more focused
- [Create a Project](https://vidbyte.pro/docs/projects/create.md): POST /api/v1/projects/create: generate a complete learning project
- [Get a Project](https://vidbyte.pro/docs/projects/get.md): GET /api/v1/projects/get/{encrypted_id}: read one project
- [List Projects](https://vidbyte.pro/docs/projects/list.md): GET /api/v1/projects/list: list the account's projects
- [Vidbyte Harnesses](https://vidbyte.pro/docs/harnesses.md): Vidbyte harnesses: hosted AI workflows with saved results and spend caps
- [Research Harness](https://vidbyte.pro/docs/harnesses/research.md): Vidbyte Research Harness: a persistent, source-grounded research portfolio
- [What Research can do](https://vidbyte.pro/docs/harnesses/research/capabilities.md): What Vidbyte Research can do: sources, limits, and the artifact it saves
- [Research threads are meant to persist](https://vidbyte.pro/docs/harnesses/research/persistent-threads.md): Vidbyte Research threads: keep the evidence collection across many runs
- [Research pricing and capacity](https://vidbyte.pro/docs/harnesses/research/pricing.md): Vidbyte Research pricing: packs, run caps, and how a run is billed
- [Research Harness API](https://vidbyte.pro/docs/harnesses/research/api.md): Research Harness API: start, extend, continue, and read research runs over HTTP
- [Create a Research run](https://vidbyte.pro/docs/harnesses/research/api/run.md): POST /api/v1/research/run: create a research thread and its first run
- [Append a run to a thread](https://vidbyte.pro/docs/harnesses/research/api/append.md): POST /api/v1/research/threads/{encrypted_id}/run: add a prompt to a research thread
- [Continue an interrupted run](https://vidbyte.pro/docs/harnesses/research/api/continue.md): POST /api/v1/research/runs/{run_id}/continue: resume an interrupted research run
- [Read run status](https://vidbyte.pro/docs/harnesses/research/api/get-run.md): GET /api/v1/research/runs/{run_id}: read a research run's status
- [List the research portfolio](https://vidbyte.pro/docs/harnesses/research/api/portfolio.md): GET /api/v1/research/portfolio: page through the account's research threads
- [Read one research thread](https://vidbyte.pro/docs/harnesses/research/api/get-thread.md): GET /api/v1/research/threads/{encrypted_id}: read one research thread
- [Start an artifact deep dive](https://vidbyte.pro/docs/harnesses/research/api/deep-dive.md): Research artifact deep dive: a browser-session route, not an API-key route

## Agents, skills, and harnesses
Ready-made agents, installable agent skills, and hosted research.

- [Vidbyte Harnesses](https://vidbyte.pro/harnesses.md): Vidbyte harnesses: hosted agent workflows paid from one prepaid balance
- [Research Harness: Hosted Deep Research With Cited, Saved Sources](https://vidbyte.pro/harnesses/research.md): Research harness pricing: Starter, Pro, Scale, and Enterprise packs, and how runs bill
- [Vidbyte Agents](https://vidbyte.pro/agents.md): Vidbyte Agents: ready-made agents run from vidbyte-cli, with a one-paste setup prompt
- [Agent Marketplace](https://vidbyte.pro/agents/marketplace.md): Vidbyte agent marketplace: every agent with its status and pricing model
- [Agent Skills](https://vidbyte.pro/agents/skills.md): Vidbyte agent skills: SKILL.md files that hand one job to a Vidbyte agent
- [Agent Dashboard](https://vidbyte.pro/agents/dashboard.md): Vidbyte agent dashboard: every run synced from vidbyte-cli, newest first (signed-in)
- [Agent Run](https://vidbyte.pro/agents/dashboard/{encrypted_id}.md): Vidbyte agent run page: one synced run's timeline, memories, files, handoff (signed-in)
- [Persistent Researcher](https://vidbyte.pro/agents/persistent-researcher.md): Persistent Researcher: hosted research that files every source in a lasting thread
- [Suggestion Agent](https://vidbyte.pro/agents/suggestion-agent.md): Suggestion Agent: ranked, critic-reviewed next actions for any goal from vidbyte-cli
- [Persistent Agent](https://vidbyte.pro/agents/persistent-agent.md): Persistent Agent: keep a local Codex session working on one task past its first answer
- [Ensemble Agent](https://vidbyte.pro/agents/ensemble-agent.md): Ensemble Agent: compare several Codex-proposed approaches before one is implemented
- [Task Board](https://vidbyte.pro/agents/task-board.md): Task Board: run a trusted task list with a fresh local Codex agent per task
- [Stages Agent](https://vidbyte.pro/agents/stages-agent.md): Stages Agent: run one job through phases, each a separate Codex agent with own settings
- [Adversarial Team](https://vidbyte.pro/agents/adversarial-team.md): Adversarial Team (planned): read-only reviewer agents that attack finished work
- [InfiniteMemory Agent](https://vidbyte.pro/agents/infinite-memory-agent.md): InfiniteMemory Agent: one long task in checked pieces, memory kept as files
- [Implement GitHub issues, one agent per issue](https://vidbyte.pro/agents/skills/github-issue-tasks.md): Implement GitHub issues, one agent per issue (Vidbyte agent skill)
- [Resolve merge conflicts on your PRs](https://vidbyte.pro/agents/skills/resolve-pr-conflicts.md): Resolve merge conflicts on your PRs (Vidbyte agent skill)
- [Research a competitor, one aspect at a time](https://vidbyte.pro/agents/skills/research-a-competitor.md): Research a competitor, one aspect at a time (Vidbyte agent skill)
- [Pick up where your agent sessions left off](https://vidbyte.pro/agents/skills/pick-up-where-my-sessions-left-off.md): Pick up where your agent sessions left off (Vidbyte agent skill)
- [Decide which pull requests to review first](https://vidbyte.pro/agents/skills/which-pr-to-review-first.md): Decide which pull requests to review first (Vidbyte agent skill)
- [Cut your coding agent's token usage](https://vidbyte.pro/agents/skills/cut-my-agent-token-usage.md): Cut your coding agent's token usage (Vidbyte agent skill)
- [Stop repeating yourself to your coding agents](https://vidbyte.pro/agents/skills/stop-repeating-yourself.md): Stop repeating yourself to your coding agents (Vidbyte agent skill)
- [Find what Slack still needs from you](https://vidbyte.pro/agents/skills/next-actions-from-slack.md): Find what Slack still needs from you (Vidbyte agent skill)
- [Turn your Notion pages into next actions](https://vidbyte.pro/agents/skills/next-actions-from-notion.md): Turn your Notion pages into next actions (Vidbyte agent skill)
- [Turn your Obsidian notes into next actions](https://vidbyte.pro/agents/skills/next-actions-from-obsidian.md): Turn your Obsidian notes into next actions (Vidbyte agent skill)
- [Turn your Google Docs into next actions](https://vidbyte.pro/agents/skills/next-actions-from-google-docs.md): Turn your Google Docs into next actions (Vidbyte agent skill)
- [Get your focus time back from too many meetings](https://vidbyte.pro/agents/skills/too-many-meetings.md): Get your focus time back from too many meetings (Vidbyte agent skill)
- [Make your vibe-coded app production ready](https://vidbyte.pro/agents/skills/vibe-coded-to-production.md): Make your vibe-coded app production ready (Vidbyte agent skill, coming soon)
- [Critique how you prompt your coding agents](https://vidbyte.pro/agents/skills/improve-my-prompting.md): Critique how you prompt your coding agents (Vidbyte agent skill, coming soon)
- [Turn the mistakes you keep correcting into agent skills](https://vidbyte.pro/agents/skills/turn-corrections-into-skills.md): Turn the mistakes you keep correcting into agent skills (Vidbyte agent skill, coming soon)
- [Find what else needs doing after your agent finishes](https://vidbyte.pro/agents/skills/what-else-needs-doing.md): Find what else needs doing after your agent finishes (Vidbyte agent skill, coming soon)
- [Predict what reviewers will say before you open a PR](https://vidbyte.pro/agents/skills/predict-the-review.md): Predict what reviewers will say before you open a PR (Vidbyte agent skill, coming soon)
- [Find the coding-agent work you never finished](https://vidbyte.pro/agents/skills/find-unfinished-agent-work.md): Find the coding-agent work you never finished (Vidbyte agent skill, coming soon)
- [Automate the chores you redo with your coding agent](https://vidbyte.pro/agents/skills/automate-recurring-chores.md): Automate the chores you redo with your coding agent (Vidbyte agent skill, coming soon)
- [Get product ideas in Slack after every merged PR](https://vidbyte.pro/agents/skills/product-ideas-after-merge.md): Get product ideas in Slack after every merged PR (Vidbyte agent skill, coming soon)
- [Turn all your notes into one list of next actions](https://vidbyte.pro/agents/skills/next-actions-from-all-my-notes.md): Turn all your notes into one list of next actions (Vidbyte agent skill, coming soon)
- [Turn your request into a checklist](https://vidbyte.pro/agents/skills/turn-my-request-into-a-checklist.md): Turn your request into a checklist (Vidbyte agent skill, coming soon)
- [Get your site cited by answer engines](https://vidbyte.pro/agents/skills/improve-my-aeo.md): Get your site cited by answer engines (Vidbyte agent skill, coming soon)
- [Convert more app users within App Review rules](https://vidbyte.pro/agents/skills/app-conversion-tactics.md): Convert more app users within App Review rules (Vidbyte agent skill, coming soon)

## Pricing, account, and company pages
Plans, API keys and balance, legal, status, and changelog.

- [Vidbyte Home](https://vidbyte.pro/index.md): Vidbyte: tools for building, running, and paying for AI agents, plus an AI learning app
- [Vidbyte Pricing](https://vidbyte.pro/pricing.md): Vidbyte learning app pricing: Explorer, Pioneer, and Enterprise plans
- [Vidbyte Changelog](https://vidbyte.pro/changelog.md): Vidbyte changelog: dated learning-app announcements, newest first
- [Vidbyte Status](https://vidbyte.pro/status.md): Vidbyte status: live health of six product areas, judged by server-error rate
- [Vidbyte Mission](https://vidbyte.pro/mission.md): Vidbyte mission: learning velocity and the active learning loop
- [Vidbyte Terms](https://vidbyte.pro/terms.md): Vidbyte Terms of Service: refunds, age, content ownership, and governing law
- [Vidbyte Privacy Policy](https://vidbyte.pro/privacy.md): Vidbyte Privacy Policy: data collected, AI providers, retention, and deletion
- [Study Evaluator](https://vidbyte.pro/study_evaluator.md): Study Evaluator: free quiz that scores your study habits on eight dimensions
- [Reasoning Evaluator](https://vidbyte.pro/reasoning.md): Reasoning Evaluator: twelve planned scenario-based reasoning exams (not live yet)
- [Feedback](https://vidbyte.pro/feedback.md): Feedback Agent preview: sample learning feedback points, no live analysis
- [Sign In](https://vidbyte.pro/login.md): Sign in to Vidbyte with Google
- [Contact Our Team](https://vidbyte.pro/contact.md): Contact the Vidbyte team: signed-in support form
- [Subscription Success](https://vidbyte.pro/success.md): Subscription success: confirmation after a Vidbyte plan checkout
- [Affiliate Program Signup](https://vidbyte.pro/affiliate_signup.md): Apply to the Vidbyte affiliate program
- [Classroom Reading Grade Level Checker](https://vidbyte.pro/classroom_reading_grading_level.md): Classroom reading grade level checker (prototype)
- [Settings](https://vidbyte.pro/settings.md): Vidbyte account settings: entry point and sections
- [Subscription Settings](https://vidbyte.pro/settings/subscription.md): Manage a Vidbyte subscription through the billing portal
- [API Settings](https://vidbyte.pro/settings/api.md): Vidbyte API settings: redirect to API keys or API usage
- [API Keys](https://vidbyte.pro/settings/api/keys.md): Create and manage Vidbyte API keys
- [API Usage](https://vidbyte.pro/settings/api/usage.md): Vidbyte API usage: see and top up the shared prepaid balance
- [Integrations Settings](https://vidbyte.pro/settings/integrations.md): Vidbyte integrations settings (turned off; redirects to Subscription)
- [Workspace Settings](https://vidbyte.pro/settings/workspace.md): Vidbyte workspace settings (turned off; redirects to Subscription)
- [Affiliate Dashboard](https://vidbyte.pro/settings/affiliate.md): Vidbyte affiliate dashboard: referral link, referrals, earnings, payouts

## Learning app pages (signed-in guides; no user data)
What each signed-in page does, and the API or CLI route to the same result.

- [The Sandbox](https://vidbyte.pro/sandbox.md): Vidbyte Sandbox: quizzes, exams, projects, and coached Q&A from your material
- [Sandbox Quiz and Exam Builder](https://vidbyte.pro/sandbox/create.md): Generate a quiz or exam from notes or documents in the Vidbyte Sandbox builder
- [Sandbox Review Vault](https://vidbyte.pro/sandbox/dashboard.md): Review Vault: reopen, retake, grade, and update saved Sandbox items
- [Project Studio](https://vidbyte.pro/sandbox/project.md): Project Studio: turn a topic into a 15-40 hour learning project delivered as a PDF
- [Project Studio Project](https://vidbyte.pro/sandbox/project/{project_id}.md): Project Studio project page: intake answers, generation progress, and the project PDF
- [Questionaire](https://vidbyte.pro/sandbox/questionaire.md): Start a Questionaire: a coached Q&A on a topic that ends in a mastery and gap report
- [Questionaire Session](https://vidbyte.pro/sandbox/questionaire/{questionaire_id}.md): Questionaire session page: answer the coach, then analyze for a mastery report
- [Questionaire Persona Library](https://vidbyte.pro/sandbox/questionaire/persona.md): Questionaire persona library: compare the seven coaching styles
- [Sandbox Quiz Generator](https://vidbyte.pro/sandbox/quiz.md): Simple Sandbox quiz form: paste material, generate, find the quiz in the Review Vault
- [Roadmap Generator](https://vidbyte.pro/roadmap/create.md): Generate a personalized learning roadmap of modules and curated resources
- [Learning Roadmap](https://vidbyte.pro/roadmap/{user_id}/{roadmap_id}.md): Learning roadmap page: syllabus, modules, resources, progress, and study calendar
- [Roadmap Resource Learn View](https://vidbyte.pro/roadmap/{user_id}/{roadmap_id}/learn/{moduleIndex}/{resourceIndex}.md): Roadmap learn view: study one resource with an assistant and practice quizzes
- [Quick Hits](https://vidbyte.pro/quickhits.md): Quick Hits: learn one topic in a 30-60 minute generated module
- [Quick Hit](https://vidbyte.pro/quickhits/{quickhit_id}.md): Saved quick hit page: pretest, eight sections, reflection, and next topics
- [Skill Learning Module](https://vidbyte.pro/m/{module_id}.md): Learning skill module page: a 7-day public link from a Vidbyte learning skill
- [Skill Item](https://vidbyte.pro/skills/{skill_id}/{encrypted_id}.md): Learning skill item page: a public 7-day view of one skill output
- [Retain Exercise](https://vidbyte.pro/skills/retain/{encrypted_id}.md): Retain exercise: a timed 15-minute recall drill from the /retain learning skill
- [Research Workspace](https://vidbyte.pro/research.md): Research workspace: run hosted research threads in the browser and read cited artifacts
- [Video Chat](https://vidbyte.pro/chat.md): Vidbyte video page: question-to-video generation, currently unavailable
- [Chat Learning Profile](https://vidbyte.pro/chat/context.md): Set your learning profile for Vidbyte video explanations
- [Chat Past Questions](https://vidbyte.pro/chat/questions.md): Past video questions: reread or copy earlier prompts
- [Socratic Tutor](https://vidbyte.pro/socratic.md): Socratic Tutor: learn a topic through up to 20 guided questions
- [Socratic Tutor Session](https://vidbyte.pro/socratic/{socratic_token}.md): Socratic Tutor session page: answer, request hints, reveal, and analyze
- [Voice Tutor](https://vidbyte.pro/tutor.md): Vidbyte Voice Tutor: spoken real-time tutoring (allowlisted accounts only)
- [Tutor Vault](https://vidbyte.pro/tutor/vault.md): Tutor Vault: search and open past Voice Tutor conversations
- [Tutor Vault Conversation](https://vidbyte.pro/tutor/vault/{conversation_token}.md): Tutor Vault conversation page: transcript, review notes, notebook, and quiz

## Machine-readable
Kept in sync with https://vidbyte.pro/robots.txt and https://vidbyte.pro/ai.txt: all three are generated from the same site configuration.

- [Full text](https://vidbyte.pro/llms-full.txt): every developer, agent, account, and app page in one file, with install, auth, and pricing steps and the agent product pages first
- [Markdown catalog](https://vidbyte.pro/api/md/_catalog): every page that has a markdown copy, with the three ways to fetch one
- [Sitemap](https://vidbyte.pro/sitemap.xml): every public HTML page with its last-modified date
- [AI usage policy (ai.txt)](https://vidbyte.pro/ai.txt): what AI systems may and may not do with Vidbyte's content, and how to cite it

## Optional
- [Vidbyte blog](https://vidbyte.pro/blog/llms.txt): Vidbyte's articles on learning science, reasoning frameworks, agent engineering, and its products.
- [Vidbyte cognitive tests](https://vidbyte.pro/games/llms.txt): Free browser-based cognitive tests and the companion articles on what each one measures.