Beatra AI Video Studio
Overview
Beatra AI Video Studio is Beatra's official agent package for short AI video
work: text-to-video, image-to-video, first/last-frame interpolation,
reference-guided generation, and editing or extending an existing clip. The
video is produced on the hosted, paid Beatra service (mcp.beatra.ai).
This catalog entry is a reviewed pointer, not the executable package. It
contains no client code and performs no Beatra operation by itself. The package
it points to bundles three standard-library Python scripts that make network
calls and store a credential. It has no self-updater: it never downloads
code and never changes its own files. Read
Install and
Security & Safety Notes before activating it.
| Pinned identity |
Value |
| Package |
beatra-ai-video-studio 1.3.5 |
| Source commit |
beatra-ai/beatra-skills@b7ab8ed (full SHA b7ab8ed02980ee44b6dac62633f8d79daf0809ab) |
| Package tree |
5183c76c894e650691c8bb638257dd202b23e66c (Git tree ID of skills/beatra-ai-video-studio at that commit) |
| License |
MIT No Attribution (MIT-0) |
The pinned tree holds regular files only (no symlinks, no executable bits, no
binaries). Confirm that with the inspect commands below.
When to Use This Skill
- Use when the user explicitly wants a short AI video clip produced on Beatra
and accepts that the work is paid and runs on a hosted service.
- Use when a still image should be animated, two frames interpolated, or an
existing clip edited or extended by Beatra.
- Do not use for local-only or offline video editing, timeline assembly, or
when the user has not agreed to send prompts and media to a third party.
Install (pinned, verified, approved twice)
Step 1: Fetch the pinned tree into a review directory
Explain that this fetches an external package from github.com, then ask
for approval. Only after approval:
set -e
umask 077
review_dir="$(mktemp -d)"
cd "$review_dir" || exit 1
git clone --quiet --filter=blob:none --no-checkout https://github.com/beatra-ai/beatra-skills.git src
git -C src checkout --quiet b7ab8ed02980ee44b6dac62633f8d79daf0809ab -- skills/beatra-ai-video-studio
test "$(git -C src rev-parse 'b7ab8ed02980ee44b6dac62633f8d79daf0809ab:skills/beatra-ai-video-studio')" = '5183c76c894e650691c8bb638257dd202b23e66c' && echo TREE-OK
Stop if the last command does not print TREE-OK. The expected commit and
tree IDs come from this catalog entry, not from the package or its website,
and Git addresses every fetched object by its hash.
Step 2: Inspect before activation
git -C src ls-tree -r b7ab8ed02980ee44b6dac62633f8d79daf0809ab -- skills/beatra-ai-video-studio | awk '$1 != "100644"' # expect no output
grep -n -E 'maybe_auto_update|download_update|_apply_update' src/skills/beatra-ai-video-studio/scripts/mcp_client.py # expect no output
The first check lists any symlink, executable, or submodule entry; the second
finds the entry points of the self-updater that earlier versions shipped.
Report to the user what the review found: SKILL.md, manifest.json,
install-source.json (the one-line marker {"schema_version":1,"source":"github"}),
the bundled Markdown references, and scripts/authorize.py,
scripts/mcp_client.py, scripts/uninstall.py (Python 3.10+, standard
library only, no dependency installation, no lifecycle hooks). Summarize the
network, credential, local-state, telemetry, and version-notice behavior
listed under Security & Safety Notes.
The manifest and pattern scans above are only inventory aids. Before activation, read every file in the pinned tree, including each script, reference and hook, and inspect all dependency and network behavior. Do not claim source review from grep output. If complete review cannot be performed, stop before copying or executing; ask for a separate review rather than treating a digest as evidence of safety.
Step 3: Copy in
Ask for a second, separate approval, because this changes agent
configuration. Then copy the reviewed tree to the host's skills directory
(Claude Code shown; use the equivalent path for other hosts):
dest="$HOME/.claude/skills/beatra-ai-video-studio"
test ! -e "$dest" && test ! -L "$dest" || { echo "destination exists; stop and ask the user"; exit 1; }
cp -R "src/skills/beatra-ai-video-studio" "$dest"
There is no update switch to turn off in this version.
Step 4: Authorize as its own decision
Authorization opens a browser sign-in and stores a bearer token. Ask first.
python3 "$dest/scripts/authorize.py"
Start a new agent session if the host discovers skills only at startup.
How It Works
- Read the installed package's
SKILL.md and its references; they define the
video routes, payload shapes, and review loop.
- Use free, non-billable discovery first (
beatra.models.list, task list) to
read current model cards, durations, resolutions, and credit estimates.
- Show the user a cost card for exactly one paid call: tool, model, duration,
resolution, estimate, and which files will be uploaded. Wait for explicit
approval. Planning or "make the clip" is not approval.
- Upload only files the user supplied and approved for this task, then submit
once with a stable
client_request_id and poll the returned task.
- Report the delivered artifact and
billing.net_charged_credits, review the
clip with the user, and ask again before any further paid stage.
Examples
Free connection check (non-billable)
python3 "$HOME/.claude/skills/beatra-ai-video-studio/scripts/mcp_client.py" verify
One approved text-to-video call
After the user approved the cost card for this exact payload:
printf '%s' '{"prompt":"A close product reveal with one slow push-in.","model":"auto","duration":5,"client_request_id":"req-7f3c2a"}' \
| python3 "$HOME/.claude/skills/beatra-ai-video-studio/scripts/mcp_client.py" call beatra.videos.generate
If the response is lost, retry only with the same client_request_id and
identical arguments; never submit a replacement while the task is running.
Update
This pinned version cannot update itself: it has no code that downloads or
replaces package files. update only reports what the store or tool that
installed the copy offers; for a copy installed by the steps above it finds
no such record and says to update the way you installed it. See the version
notice under Security & Safety Notes.
To move to a newer version, repeat Steps 1 to 3 for that commit in a new
review directory, show the user the file-level changes against the installed
copy (for example git -C src diff --stat b7ab8ed <new-commit> -- skills/beatra-ai-video-studio),
and replace the directory only after approval. A version not listed in this
entry has not been reviewed here.
Uninstall
dest="$HOME/.claude/skills/beatra-ai-video-studio"
python3 "$dest/scripts/uninstall.py" --dry-run
python3 "$dest/scripts/uninstall.py"
Read the JSON decision. disconnected means this was the last Beatra skill on
the device: the token was revoked (when revoked is true) and the files
credentials.json, installation.json, host.json, skills.json,
network.json, balance-holds.json, pending_authorization.json, and
.authorize.lock under ~/.beatra were removed. keep_connection means
another Beatra skill still uses the shared token, which stays.
revoke_retry means the server refused the revocation and nothing changed;
run it again shortly. The script never deletes the package directory: after
confirming the path it prints with the user, delete $dest.
uninstall.py does not remove ~/.beatra/updates/. When the decision is
disconnected and no other Beatra skill is installed, remove that directory
and then the empty ~/.beatra; otherwise leave them. A token that could not
be revoked expires after 15 days idle and can be revoked at once in the
Beatra Console under Agents.
Best Practices
- ✅ Keep one approval per paid call, per upload set, and per install step.
- ✅ Start with the shortest duration and lowest resolution the model admits.
- ✅ Report net charged credits from the terminal task, not the estimate.
- ❌ Do not replace the installed copy with an unreviewed version silently.
- ❌ Do not print, paste, or move the token out of
~/.beatra/credentials.json.
- ❌ Do not retry a paid call with a new request ID after an uncertain response.
Limitations
- Requires a Beatra account, network access, Python 3.10+, and prepaid credits;
the sign-up gift usually cannot cover a video generation.
- Short clips only; not a timeline editor. Multi-shot work is delivered as
separate clips.
- Generation is asynchronous and can take minutes. Upload size is capped at
100 MB and a model may impose a lower limit.
- Models, prices, schemas, and availability are controlled by Beatra and its
upstream model providers and can change; the service is provided as is.
- Outputs are probabilistic, not guaranteed unique or non-infringing, and need
human review before publication.
Security & Safety Notes
risk: critical is deliberate.
- Network. Every request is HTTPS and refuses redirects. Authorization
uses
api.beatra.ai and tool calls go to mcp.beatra.ai; an upload is a
PUT of the approved file to the HTTPS address the server returns, without
the bearer token. Unless a cached decision is still valid (10 minutes to 7
days), each run first asks https://api.beatra.ai/skill-route, and also
https://api.beatra.cn/skill-route when the first is slow or answers cn,
which network to use. These probes carry no credential or package identity.
When the answer is cn, authorization and tool calls go to api.beatra.cn
and mcp.beatra.cn with the same bearer token, and the version notice
reads from www.beatra.cn; Beatra serves these .cn hosts from mainland
China.
python3 "$dest/scripts/mcp_client.py" network --use ai pins the .ai
hosts and stops the probes; a failed version-notice read can still retry
once on the other network.
- Version notice, no self-update. This version has no code that downloads
or replaces package files and writes nothing inside the install directory.
Before
verify, tools, upload, or call, at most once every 24 hours
per install path, it records the check time and looks, read-only, for
records another installer left beside the package (ClawHub origin.json,
SkillHub or skills CLI lock entries, gh skill frontmatter). A copy
installed by the steps above normally has none and makes no request. When
one is found, it reads https://beatra.ai/skills/beatra-ai-video-studio/sources.json
(no credential, 3-second timeout, 64 KiB cap) and, if that installer offers a
newer version, prints one line to stderr with the installer's own update
command. It never runs the command.
- Uploads and third parties. Prompts and uploaded media are sent to Beatra
and forwarded to the third-party model provider executing the task. Beatra
runs the service outside mainland China; on the
cn route (see Network)
requests reach it through Beatra's hosts in mainland China. See the
privacy policy.
- Billing. Paid tools consume prepaid credits. Credits for tasks that fail
for reasons attributable to Beatra or the upstream model are returned
automatically; otherwise purchases are non-refundable, and actual usage can
exceed the estimate and leave a negative balance. See the
terms.
- Credentials.
authorize.py runs an OAuth device flow, opens the browser
on the approval page, and stores one bearer token in
~/.beatra/credentials.json (directory 0700, file 0600). The token never
goes into argv, environment variables, or output. It carries full Beatra
scope, including wallet:spend, has a sliding 15-day idle lifetime, and is
shared by every Beatra skill on the device. While the user has not yet
selected Allow, the pending device code is kept in
~/.beatra/pending_authorization.json (0600) so a rerun finishes the same
approval instead of opening a new one. When verify, tools, upload, or
call finds no working token, the client makes one check of that saved
approval and stores the token if the user has already selected Allow; it
never starts a new authorization or opens a browser.
- Local state.
~/.beatra/ also holds installation.json (random
installation reference), host.json (agent platform and hostname),
skills.json (installed Beatra skill paths), network.json (route
decision, with a country code when the route answer includes one),
balance-holds.json (after an insufficient-credit refusal, that tool is
answered locally for up to 30 minutes; call --recheck-balance sends one
call), a short-lived .authorize.lock, and, per install path,
updates/<id>/state.json (last version check), source.json (install
source), and registration.json (last registration). Files are 0600
except host.json, which follows the umask.
- Telemetry. Authorization sends the hostname as the device name, the
package slug and version, the detected agent platform (for example
claude-code, from environment variables), and the installation reference.
After connecting, at most once a day unless the version, platform, route, or
install source changes, the client makes a non-billable
beatra.installations.register call with those package and platform
fields, the release channel, the network route, the install source
(github for this tree), and coarse runtime facts: OS and OS version, CPU
architecture, Python version, locale, and UTC offset. Every tool call adds
source_package_slug and source_platform. There is no opt-out for these
fields.
- Retention. Account data is kept for the life of the account. Task
records, artifacts, and billing ledgers are retained after account closure
for audit and tax purposes.
- Content rights. Users keep rights in their inputs and receive Beatra's
rights in outputs; they must hold rights to uploaded media and must not
create deceptive impersonations.
Additional Resources
1---2name: beatra-ai-video-studio3description: "Install and use the official Beatra AI Video Studio package, pinned by digest, for paid text-to-video, image-to-video, and video edit or extend jobs on the hosted Beatra service."4category: media5risk: critical6source: community7source_repo: beatra-ai/beatra-skills8source_type: official9date_added: "2026-09-17"10author: beatra-ai11tags: [video-generation, text-to-video, image-to-video, video-editing, mcp, paid-api, beatra]12tools: [claude, codex, cursor, gemini]13license: "MIT-0"14license_source: "https://github.com/beatra-ai/beatra-skills/blob/b7ab8ed02980ee44b6dac62633f8d79daf0809ab/LICENSE"15metadata:16 source: https://skillmd.com/skills/iradoweck/beatra-ai-video-studio17---1819# Beatra AI Video Studio2021## Overview2223Beatra AI Video Studio is Beatra's official agent package for short AI video24work: text-to-video, image-to-video, first/last-frame interpolation,25reference-guided generation, and editing or extending an existing clip. The26video is produced on the hosted, paid Beatra service (`mcp.beatra.ai`).2728This catalog entry is a **reviewed pointer, not the executable package**. It29contains no client code and performs no Beatra operation by itself. The package30it points to bundles three standard-library Python scripts that make network31calls and store a credential. It has no self-updater: it never downloads32code and never changes its own files. Read33[Install](#install-pinned-verified-approved-twice) and34[Security & Safety Notes](#security--safety-notes) before activating it.3536| Pinned identity | Value |37| --- | --- |38| Package | `beatra-ai-video-studio` `1.3.5` |39| Source commit | [`beatra-ai/beatra-skills@b7ab8ed`](https://github.com/beatra-ai/beatra-skills/tree/b7ab8ed02980ee44b6dac62633f8d79daf0809ab/skills/beatra-ai-video-studio) (full SHA `b7ab8ed02980ee44b6dac62633f8d79daf0809ab`) |40| Package tree | `5183c76c894e650691c8bb638257dd202b23e66c` (Git tree ID of `skills/beatra-ai-video-studio` at that commit) |41| License | MIT No Attribution (MIT-0) |4243The pinned tree holds regular files only (no symlinks, no executable bits, no44binaries). Confirm that with the inspect commands below.4546## When to Use This Skill4748- Use when the user explicitly wants a short AI video clip produced on Beatra49 and accepts that the work is paid and runs on a hosted service.50- Use when a still image should be animated, two frames interpolated, or an51 existing clip edited or extended by Beatra.52- Do not use for local-only or offline video editing, timeline assembly, or53 when the user has not agreed to send prompts and media to a third party.5455## Install (pinned, verified, approved twice)5657### Step 1: Fetch the pinned tree into a review directory5859Explain that this fetches an external package from `github.com`, then ask60for approval. Only after approval:6162```bash63set -e64umask 07765review_dir="$(mktemp -d)"66cd "$review_dir" || exit 167git clone --quiet --filter=blob:none --no-checkout https://github.com/beatra-ai/beatra-skills.git src68git -C src checkout --quiet b7ab8ed02980ee44b6dac62633f8d79daf0809ab -- skills/beatra-ai-video-studio69test "$(git -C src rev-parse 'b7ab8ed02980ee44b6dac62633f8d79daf0809ab:skills/beatra-ai-video-studio')" = '5183c76c894e650691c8bb638257dd202b23e66c' && echo TREE-OK70```7172Stop if the last command does not print `TREE-OK`. The expected commit and73tree IDs come from this catalog entry, not from the package or its website,74and Git addresses every fetched object by its hash.7576### Step 2: Inspect before activation7778```bash79git -C src ls-tree -r b7ab8ed02980ee44b6dac62633f8d79daf0809ab -- skills/beatra-ai-video-studio | awk '$1 != "100644"' # expect no output80grep -n -E 'maybe_auto_update|download_update|_apply_update' src/skills/beatra-ai-video-studio/scripts/mcp_client.py # expect no output81```8283The first check lists any symlink, executable, or submodule entry; the second84finds the entry points of the self-updater that earlier versions shipped.8586Report to the user what the review found: `SKILL.md`, `manifest.json`,87`install-source.json` (the one-line marker `{"schema_version":1,"source":"github"}`),88the bundled Markdown references, and `scripts/authorize.py`,89`scripts/mcp_client.py`, `scripts/uninstall.py` (Python 3.10+, standard90library only, no dependency installation, no lifecycle hooks). Summarize the91network, credential, local-state, telemetry, and version-notice behavior92listed under [Security & Safety Notes](#security--safety-notes).9394The manifest and pattern scans above are only inventory aids. Before activation, read every file in the pinned tree, including each script, reference and hook, and inspect all dependency and network behavior. Do not claim source review from grep output. If complete review cannot be performed, stop before copying or executing; ask for a separate review rather than treating a digest as evidence of safety.9596### Step 3: Copy in9798Ask for a **second, separate** approval, because this changes agent99configuration. Then copy the reviewed tree to the host's skills directory100(Claude Code shown; use the equivalent path for other hosts):101102```bash103dest="$HOME/.claude/skills/beatra-ai-video-studio"104test ! -e "$dest" && test ! -L "$dest" || { echo "destination exists; stop and ask the user"; exit 1; }105cp -R "src/skills/beatra-ai-video-studio" "$dest"106```107108There is no update switch to turn off in this version.109110### Step 4: Authorize as its own decision111112Authorization opens a browser sign-in and stores a bearer token. Ask first.113114```bash115python3 "$dest/scripts/authorize.py"116```117118Start a new agent session if the host discovers skills only at startup.119120## How It Works1211221. Read the installed package's `SKILL.md` and its references; they define the123 video routes, payload shapes, and review loop.1242. Use free, non-billable discovery first (`beatra.models.list`, task list) to125 read current model cards, durations, resolutions, and credit estimates.1263. Show the user a cost card for exactly one paid call: tool, model, duration,127 resolution, estimate, and which files will be uploaded. Wait for explicit128 approval. Planning or "make the clip" is not approval.1294. Upload only files the user supplied and approved for this task, then submit130 once with a stable `client_request_id` and poll the returned task.1315. Report the delivered artifact and `billing.net_charged_credits`, review the132 clip with the user, and ask again before any further paid stage.133134## Examples135136### Free connection check (non-billable)137138```bash139python3 "$HOME/.claude/skills/beatra-ai-video-studio/scripts/mcp_client.py" verify140```141142### One approved text-to-video call143144After the user approved the cost card for this exact payload:145146```bash147printf '%s' '{"prompt":"A close product reveal with one slow push-in.","model":"auto","duration":5,"client_request_id":"req-7f3c2a"}' \148 | python3 "$HOME/.claude/skills/beatra-ai-video-studio/scripts/mcp_client.py" call beatra.videos.generate149```150151If the response is lost, retry only with the same `client_request_id` and152identical arguments; never submit a replacement while the task is running.153154## Update155156This pinned version cannot update itself: it has no code that downloads or157replaces package files. `update` only reports what the store or tool that158installed the copy offers; for a copy installed by the steps above it finds159no such record and says to update the way you installed it. See the version160notice under [Security & Safety Notes](#security--safety-notes).161162To move to a newer version, repeat Steps 1 to 3 for that commit in a new163review directory, show the user the file-level changes against the installed164copy (for example `git -C src diff --stat b7ab8ed <new-commit> -- skills/beatra-ai-video-studio`),165and replace the directory only after approval. A version not listed in this166entry has not been reviewed here.167168## Uninstall169170```bash171dest="$HOME/.claude/skills/beatra-ai-video-studio"172python3 "$dest/scripts/uninstall.py" --dry-run173python3 "$dest/scripts/uninstall.py"174```175176Read the JSON decision. `disconnected` means this was the last Beatra skill on177the device: the token was revoked (when `revoked` is `true`) and the files178`credentials.json`, `installation.json`, `host.json`, `skills.json`,179`network.json`, `balance-holds.json`, `pending_authorization.json`, and180`.authorize.lock` under `~/.beatra` were removed. `keep_connection` means181another Beatra skill still uses the shared token, which stays.182`revoke_retry` means the server refused the revocation and nothing changed;183run it again shortly. The script never deletes the package directory: after184confirming the path it prints with the user, delete `$dest`.185186`uninstall.py` does not remove `~/.beatra/updates/`. When the decision is187`disconnected` and no other Beatra skill is installed, remove that directory188and then the empty `~/.beatra`; otherwise leave them. A token that could not189be revoked expires after 15 days idle and can be revoked at once in the190Beatra Console under Agents.191192## Best Practices193194- ✅ Keep one approval per paid call, per upload set, and per install step.195- ✅ Start with the shortest duration and lowest resolution the model admits.196- ✅ Report net charged credits from the terminal task, not the estimate.197- ❌ Do not replace the installed copy with an unreviewed version silently.198- ❌ Do not print, paste, or move the token out of `~/.beatra/credentials.json`.199- ❌ Do not retry a paid call with a new request ID after an uncertain response.200201## Limitations202203- Requires a Beatra account, network access, Python 3.10+, and prepaid credits;204 the sign-up gift usually cannot cover a video generation.205- Short clips only; not a timeline editor. Multi-shot work is delivered as206 separate clips.207- Generation is asynchronous and can take minutes. Upload size is capped at208 100 MB and a model may impose a lower limit.209- Models, prices, schemas, and availability are controlled by Beatra and its210 upstream model providers and can change; the service is provided as is.211- Outputs are probabilistic, not guaranteed unique or non-infringing, and need212 human review before publication.213214## Security & Safety Notes215216`risk: critical` is deliberate.217218- **Network.** Every request is HTTPS and refuses redirects. Authorization219 uses `api.beatra.ai` and tool calls go to `mcp.beatra.ai`; an upload is a220 `PUT` of the approved file to the HTTPS address the server returns, without221 the bearer token. Unless a cached decision is still valid (10 minutes to 7222 days), each run first asks `https://api.beatra.ai/skill-route`, and also223 `https://api.beatra.cn/skill-route` when the first is slow or answers `cn`,224 which network to use. These probes carry no credential or package identity.225 When the answer is `cn`, authorization and tool calls go to `api.beatra.cn`226 and `mcp.beatra.cn` with the same bearer token, and the version notice227 reads from `www.beatra.cn`; Beatra serves these `.cn` hosts from mainland228 China.229 `python3 "$dest/scripts/mcp_client.py" network --use ai` pins the `.ai`230 hosts and stops the probes; a failed version-notice read can still retry231 once on the other network.232- **Version notice, no self-update.** This version has no code that downloads233 or replaces package files and writes nothing inside the install directory.234 Before `verify`, `tools`, `upload`, or `call`, at most once every 24 hours235 per install path, it records the check time and looks, read-only, for236 records another installer left beside the package (ClawHub `origin.json`,237 SkillHub or `skills` CLI lock entries, `gh skill` frontmatter). A copy238 installed by the steps above normally has none and makes no request. When239 one is found, it reads `https://beatra.ai/skills/beatra-ai-video-studio/sources.json`240 (no credential, 3-second timeout, 64 KiB cap) and, if that installer offers a241 newer version, prints one line to stderr with the installer's own update242 command. It never runs the command.243- **Uploads and third parties.** Prompts and uploaded media are sent to Beatra244 and forwarded to the third-party model provider executing the task. Beatra245 runs the service outside mainland China; on the `cn` route (see Network)246 requests reach it through Beatra's hosts in mainland China. See the247 [privacy policy](https://beatra.ai/en/privacy).248- **Billing.** Paid tools consume prepaid credits. Credits for tasks that fail249 for reasons attributable to Beatra or the upstream model are returned250 automatically; otherwise purchases are non-refundable, and actual usage can251 exceed the estimate and leave a negative balance. See the252 [terms](https://beatra.ai/en/terms).253- **Credentials.** `authorize.py` runs an OAuth device flow, opens the browser254 on the approval page, and stores one bearer token in255 `~/.beatra/credentials.json` (directory `0700`, file `0600`). The token never256 goes into argv, environment variables, or output. It carries full Beatra257 scope, including `wallet:spend`, has a sliding 15-day idle lifetime, and is258 shared by every Beatra skill on the device. While the user has not yet259 selected Allow, the pending device code is kept in260 `~/.beatra/pending_authorization.json` (`0600`) so a rerun finishes the same261 approval instead of opening a new one. When `verify`, `tools`, `upload`, or262 `call` finds no working token, the client makes one check of that saved263 approval and stores the token if the user has already selected Allow; it264 never starts a new authorization or opens a browser.265- **Local state.** `~/.beatra/` also holds `installation.json` (random266 installation reference), `host.json` (agent platform and hostname),267 `skills.json` (installed Beatra skill paths), `network.json` (route268 decision, with a country code when the route answer includes one),269 `balance-holds.json` (after an insufficient-credit refusal, that tool is270 answered locally for up to 30 minutes; `call --recheck-balance` sends one271 call), a short-lived `.authorize.lock`, and, per install path,272 `updates/<id>/state.json` (last version check), `source.json` (install273 source), and `registration.json` (last registration). Files are `0600`274 except `host.json`, which follows the umask.275- **Telemetry.** Authorization sends the hostname as the device name, the276 package slug and version, the detected agent platform (for example277 `claude-code`, from environment variables), and the installation reference.278 After connecting, at most once a day unless the version, platform, route, or279 install source changes, the client makes a non-billable280 `beatra.installations.register` call with those package and platform281 fields, the release channel, the network route, the install source282 (`github` for this tree), and coarse runtime facts: OS and OS version, CPU283 architecture, Python version, locale, and UTC offset. Every tool call adds284 `source_package_slug` and `source_platform`. There is no opt-out for these285 fields.286- **Retention.** Account data is kept for the life of the account. Task287 records, artifacts, and billing ledgers are retained after account closure288 for audit and tax purposes.289- **Content rights.** Users keep rights in their inputs and receive Beatra's290 rights in outputs; they must hold rights to uploaded media and must not291 create deceptive impersonations.292293## Additional Resources294295- [Source tree at the pinned commit](https://github.com/beatra-ai/beatra-skills/tree/b7ab8ed02980ee44b6dac62633f8d79daf0809ab/skills/beatra-ai-video-studio)296- [Product page](https://beatra.ai/skills/beatra-ai-video-studio)297- [Terms](https://beatra.ai/en/terms) and [privacy policy](https://beatra.ai/en/privacy)