Scout Launch Status

Scout beta release is ready with limits.

This page mirrors the launch readiness contract in plain language. It exists so beta testers and operators can see what is verified, which release gates are closed, and which future public self-serve capabilities remain intentionally deferred.

Current verdict

Readiness is intentionally split.

Beta

Controlled beta: ready_with_limits

Controlled beta can continue for approved testers using local install, Docker-from-source, skill usage, and approved hosted beta keys.

Release

Launch readiness: ready

The release checklist has no current blockers after the artifact-only beta tag and downloaded-artifact smoke.

Boundary

Public self-serve is deferred

PyPI, GHCR, Docker Hub, paid checkout, security-clean claims, and unlimited hosted crawling are not part of this beta release.

Owner summary

What changed in the blocker burndown?

Arijit decisions: closed

Pricing posture, artifact-only publishing, Docker publishing deferral, and Crawl4AI/lxml private-beta exception are recorded.

Codex gates: closed

The real `v0.1.0-beta.1` release workflow passed and the downloaded artifacts were smoke-tested locally.

Stripe paid checkout: deferred

Manual hosted beta key provisioning remains the current path. Paid checkout must reopen before self-serve billing is enabled.

Current blockers: 0

`scout launch-readiness --json` reports no current release blockers.

Decision packet for Arijit

Decisions recorded for the beta path.

These decisions unlock packaging work without approving public self-serve SaaS claims.

License: Apache-2.0 for Scout local/core

Closed. Aligns with Crawl4AI's Apache-2.0 posture and keeps local adoption straightforward.

Pricing: controlled beta posture

Rejects arbitrary placeholder pricing, keeps hosted beta metered, and requires cost/volume/margin modeling before self-serve paid launch.

Publishing: artifact-only private-beta v* tag first

Ship downloadable GitHub Release artifacts after license files land; defer PyPI, GHCR, and Docker Hub.

Crawl4AI/lxml: private-beta exception only

Private-beta exception approved; public registry and security-clean claims still require a clean audit or separate exception.

Hosted: invite-only keys

Hosted beta can continue with approved testers and finite credits; paid self-serve checkout is deferred.

Blocker summary

What kind of work remains?

Current blockers: 0

The executable readiness checker has no remaining current beta release blockers.

Future public registry work

PyPI, GHCR, and Docker Hub remain future decisions, not current beta blockers.

Future paid checkout work

Real Stripe test-mode checkout and webhook smoke must happen before paid self-serve checkout is enabled.

Future security-clean work

The dependency audit remains visible; public security-clean claims require an upstream fix, replacement path, or separate public exception.

Blocker keys

Use the same IDs as the readiness report.

The prior blocker keys are now closed or explicitly deferred out of the controlled beta release scope.

Founder / risk decisions

public-pricing-and-hosted-usage-limits, registry-publishing-policy, docker-image-publishing-policy, and crawl4ai-lxml-risk-decision are recorded as completed decision records.

Codex implementation gates

github-release-workflow-run and release-artifact-smoke are closed by the `v0.1.0-beta.1` release evidence.

External smoke

stripe-real-test-mode-smoke is deferred until paid self-serve checkout enters scope.

Executable evidence

Run the readiness checker locally.

scout launch-readiness
scout launch-readiness --json
scout launch-readiness --owner Arijit
scout launch-readiness --blocker-type engineering
scout launch-readiness --blocker-id public-pricing-and-hosted-usage-limits
scout launch-readiness --require-public

Filtered reports

`--owner`, `--blocker-type`, and `--blocker-id` change the displayed blocker list only. They do not change the underlying public launch verdict.

Release gate

`--require-public` is expected to pass for the current controlled beta release after the blocker burndown.

Decision workflow

Turn blocker keys into reviewable founder decisions.

Use the installed Scout CLI to generate draft decision records, edit them into completed decisions, and validate them before any launch gate treats them as evidence.

scout launch-decision-drafts \
  --owner Arijit \
  --include-shared-owner \
  --decision-date YYYYMMDD

scout launch-decision-draft \
  --blocker-id public-pricing-and-hosted-usage-limits \
  --decision-id SCOUT-DEC-YYYYMMDD-NN

scout launch-decision-check --check-existing
scout launch-decision-check --check-drafts

Drafts are not approvals

Drafts live under docs/product/founder-decision-drafts/. They do not count as launch evidence until reviewed, moved into the completed record naming pattern, and validated. Run scout launch-decision-check --check-drafts to confirm they are still non-approving review aids.

Packet first, single draft when needed

The bulk command prepares Arijit-owned and shared decision drafts. Use the single-draft command when one blocker needs a separate discussion.

Review the draft first

Do not edit the release checklist from a draft alone. Decide whether the recommendation is approved, rejected, or still deferred.

Move approved content into a record

Move approved content into docs/product/founder-decision-record-SCOUT-DEC-YYYYMMDD-NN.md, replace every placeholder, and keep public launch blocked unless explicitly approved.

Validate before launch gates change

Only after validation passes should Codex update the release checklist, evidence index, license files, package metadata, release tags, or payment smoke evidence.