Launchfiles logoLaunchfiles

EDUCATION · 3 MINUTES · NO SIGN-UP

See a Launchfiles packet in 3 minutes.

Walk a sample mission from ask → bound → result. Nothing runs live. Nothing publishes. You leave knowing what a receipt looks like.

  • 1 AskAsk what you want
  • 2. BoundSee scope + limits
  • 3. ResultInspect the packet

Sample only · No card · No auto-publish · No credentials on this page

WHAT YOU'LL SEE

What you'll see in the sample

1 Ask

The request and what is off-limits

2 Bound

Scope, limits, and forbidden actions

3 Result

Packet: artifact, gates, next safe action

SAMPLELF-SAMPLE-PACKET
Bound scope: 1 route inspection, read-only proof
Deterministic gates: file scope, tests, types, security passed
Signed receipt: model invocation traces + hash verification
Founder verdict: human approval required before any merge or deploy

Then pick ONE next step

1

See a sample

Public quickstart walkthrough without signing up. Education only — nothing runs, no publish.

See sample
2

Try a free page radar

Paste a public URL for a real first-slice packet. Free tools run on public URLs without an account.

Try page radar

Join beta waitlist

Request invited workspace capacity when you want to run missions on your company.

Join waitlist

Other paths: pilot · workspace beta

Acquisition commercial truth

Education only — nothing runs live, no auto-publish. Workspace credit packs and product standalones are available on Pricing. This page does not complete a purchase.

Understand

Public = understand

Sample loop and receipt shape only. Nothing runs live; zero autonomous publish or deploy.

Private

Bound = private

Challenge credentials never appear in public HTML. Empty evidence stays honestly blocked.

Credits

Workspace credits

Workspace credit packs can be bought on Pricing. Credits settle after the provider webhook — a return URL is not settlement proof. Buying credits is not a workspace invite. See Pricing.

Products

Product standalones

Product standalones have their own public subscribe routes. No waitlist and no invited workspace required. See Pricing.

Join the waitlist

Interest only — no auto-approval. We invite founders in batches. Private beta access is still founder-gated.

For authenticated agents (advanced 9-step challenge path)9-step challenge

Use these steps only inside the scoped challenge manifest. A green status is not execution proof; completion needs the right receipt class and terminal state.

  1. Read the manifest

    Use the authenticated challenge manifest as the callable contract. Public discovery explains boundaries; it does not grant execution authority.

  2. Open private challenge access

    Start from an authenticated founder session or a private challenge handoff. Public pages and shared artifacts never display access credentials.

  3. Create product context

    Create a temporary challenge product only, scoped to one product family (launchfiles for this public path) and named with the permitted challenge prefix. The authenticated manifest carries the exact naming rule; it is never published here.

  4. Submit a bounded mission

    Send one mission with an explicit product, scope, forbidden actions, and expected artifact or blocker output.

  5. Poll mission status

    Poll the returned URL first when present. Treat accepted, queued, and running states as progress, not completion. Stalled or no-progress is not a receipt — re-scope or wait; never claim done without a terminal receipt.

  6. Retrieve the artifact

    Inspect the artifact, handoff, action register, gates, warnings, and side-effect boundary before making any claim.

  7. Inspect the model receipt

    A model receipt proves a bounded model invocation and its hashes or provider trace. It is not the same as lifecycle execution.

  8. Verify no side effects

    Confirm that no publish, deploy, outbound send, spend, billing mutation, customer-state change, or real-product mutation occurred.

  9. Clean up temporary state

    Delete the challenge product when the mission is complete or blocked, then report receipt truth and remaining blockers.

Four-part explanation

See why each mission phase unlocks the next.

The motion follows a real change in product state. It starts only when you ask for it, runs once, and ends at the founder decision.

What changes

The objective, permitted product context, forbidden actions, and expected result become explicit.

Why the next phase is allowed

The system may run only after it can tell useful work from out-of-scope work.

ResultA mission that is safe to start and possible to evaluate.

What founders run after access

The read-only challenge proves the loop. Inside a workspace, /today exposes page repair, design-system checks, and landing briefs as bounded jobs — still no autonomous publish or deploy.

  • Page repair package. Bound a public route, rank copy/CTA repairs, attach acceptance checks, and keep GSC volume honest (empty means not invented).
  • Design-system check. Score hierarchy and brand fit against the product design_system. Screenshot + DOM evidence required before any compliance claim.
  • Landing page brief. Draft a no-publish page-build brief: sections, primary CTA, secondary path, and forbidden outcome claims.

After beta access, open /today and pick one job chip, or explore /use-from-agent.

Approval boundary

Approval means the founder allowed one exact next action inside one scope. It does not approve publish, deploy, spend, billing, outbound send, customer mutation, or durable memory interpretation unless that action is explicitly named.

Private access stays private.

This public quickstart never displays challenge credentials. Authentication grants bounded access; it does not grant approval for publish, deploy, spend, or customer mutation.

Receipt example shape

A model receipt proves the bounded model call. Lifecycle execution requires a separate receipt when bounded work actually runs and verification passes.

receipt_id
Unique durable receipt identifier.
receipt_class
model_invocation, artifact_verification, handoff, lifecycle_execution, or side_effect.
work_order_id
The exact bounded work order the receipt belongs to.
status
completed, blocked, failed, skipped, or not_run.
verification_checks
Hashes, provider trace, gate output, artifact refs, and side-effect truth.
terminal_state
The final state that may be reported without inference.

What is not allowed

  • Treating this education page as a purchase flow (workspace credit packs and product standalones are available on Pricing; this page does not complete a purchase)
  • Autonomous execution outside the approved challenge scope
  • Publish, deploy, outbound send, spend, billing mutation, or customer-state mutation from a challenge run
  • Treating a model response, status color, action card, or handoff as an execution receipt
  • Using public OpenAPI, local repository files, shell scripts, databases, or logs as production challenge authority
  • Displaying or inventing challenge access credentials on this public page