agentleFS
Sign inStart free
lorena-bordonaba-paulorena-bordonaba-pau/youtube-agent/llms.txt

youtube-agent

llms.txt2 starsChanged 47 days agoMITView on GitHub

A Claude Code agent that analyses a YouTube channel with real API data and labels every figure with its provenance, so an estimate can never be presented as a measurement. The defining idea: every tool returns a source field — youtube_api (measured), derived (computed), heuristic (own rubric), config (local file) or generated (model artefact). The operating contract in CLAUDE.md forces the agent to respect it when citing. Tools that cannot measure something return the open question instead of a plausible…

  • Reads credentials

The file

Create a free account to mount
Files
llms.txt
# youtube-agent

> A Claude Code agent that analyses a YouTube channel with real API data and
> labels every figure with its provenance, so an estimate can never be
> presented as a measurement.

The defining idea: every tool returns a `source` field — `youtube_api`
(measured), `derived` (computed), `heuristic` (own rubric), `config` (local
file) or `generated` (model artefact). The operating contract in CLAUDE.md
forces the agent to respect it when citing. Tools that cannot measure something
return the open question instead of a plausible number.

Needs Python 3.10+. A `YOUTUBE_API_KEY` unlocks all public-data tools; OAuth is
only required for the channel owner's private analytics.

## Docs

- [README](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/README.md): what it is, what you can ask it, how it works
- [Install guide](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/docs/install.md): step by step, written for an agent to follow
- [Operating contract](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/CLAUDE.md): identity, honesty rules, transparency protocol
- [Tool catalogue](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/TOOLS.md): all 31 tools, their commands and quota cost
- [Limits](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/LIMITS.md): what it cannot do and why
- [Troubleshooting](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/docs/troubleshooting.md): exit codes and known failure modes
- [Contributing](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/CONTRIBUTING.md): the one non-negotiable rule
- [Spanish README](https://github.com/lorena-bordonaba-pau/youtube-agent/blob/main/docs/README_es.md)

## Optional

- 23 data tools: analytics, retention curve, traffic, outliers, keywords, transcripts
- 4 image tools: provider-agnostic generation, editing, deterministic export, channel packaging
- A guided start: /kickoff builds the strategy, /competitors maps the market in any language with the creator validating each channel
- 22 skills with fixed tool execution orders, including one that forbids answering about itself from memory
- A strategist subagent for content pillars, and hooks that ask before spending credits or quota
- Scoring rubrics that ship uncalibrated on purpose, to be built from your own channel

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free