agentleFS
butterbase-aibutterbase-ai/butterbase-plugin/skills/journey-submit/SKILL.md

journey-submit

Skill535 starsChanged 5 months agoMITView on GitHub

Use as the final stage of the Butterbase journey when hackathon_mode is true and journey-deploy has passed. Resolves which hackathon to submit to (asking the user when multiple are open), walks every field in the hackathon's returned field_schema with the user one at a time, then calls prep_and_submit_hackathon_entry. Writes the receipt to docs/butterbase/05-submission.md.

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If docs/butterbase/03-preflight.md is missing, older than 24 hours, or 00-state.md has app_id: null, invoke …

    line 18
  2. Very specific:

    Always start with action: "prep".

    line 33
  3. Very specific:

    Use next_call.arguments.data from the prep response as the literal template — it has one key per field with a placeholder string.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Journey: Hackathon submission

Stage 5 (final) of the guided journey. Resolve the active hackathon, confirm fields, submit.

When to use

  • Dispatched by journey when current_stage: submit and hackathon_mode: true.
  • Directly via /butterbase-skills:submit.
  • No-op (returns with "submit is disabled outside hackathon mode") if hackathon_mode: false.

Preflight

If docs/butterbase/03-preflight.md is missing, older than 24 hours, or 00-state.md has app_id: null, invoke butterbase-skills:journey-preflight first. Wait for it to return successfully before proceeding.

Additionally: refuse to run unless deploy is ticked in 00-state.md. If it is not, tell the user to run /butterbase-skills:journey-deploy first.

Inputs

  • docs/butterbase/01-idea.md — title / tagline candidates.
  • docs/butterbase/02-plan.md — feature list.
  • docs/butterbase/04-build-log.md — what actually shipped.
  • docs/butterbase/00-state.md — app_id, deployed_url.

Procedure

Step 1 — Prep (resolve which hackathon)

Always start with action: "prep". Do not assume there is exactly one hackathon, and do not invent a field schema from prior context — the platform owns the schema and it varies per hackathon.

// First, ask the user: "Do you have a submission code from the organizer? (paste it, or 'no')"
// If yes, pass it. If no, omit it.
{
  "tool": "prep_and_submit_hackathon_entry",
  "arguments": {
    "action": "prep",
    "submission_code": "<optional, from user>"
  }
}

The response shape is:

{
  "matched": null | { "slug", "name", "submission_deadline", "ends_at", "field_schema": { "fields": [...] } },
  "match_reason": null | "submission_code" | "already_bound" | "single_open",
  "open_hackathons": [{ "slug", "name", "starts_at", "ends_at", "submission_deadline" }, ...],
  "next_call": { ... }   // present only when matched is non-null
}

Branch on the result:

matchedopen_hackathons.lengthAction
non-nullanyResolved — continue to Step 2 with matched.field_schema. Mention match_reason to the user ("Resolved via your submission code", "You're already a participant in X", "Only one hackathon is open: X").
null0No hackathon is currently open for submissions. Tell the user, stop, and do not tick submit.
null1One hackathon is open but the user isn't bound and didn't supply a code. Ask: "The only open hackathon is '<name>' (deadline <submission_deadline>). Submit to this one? If so, paste the submission_code the organizer gave you." Re-run prep with the code.
null≥ 2Multiple open hackathons — ask the user which one. Present the list (name, slug, deadline) and ask: "Which hackathon are you submitting to? Paste its submission_code." Re-run prep with the code. Never guess.

Anti-pattern: do not re-run prep in a loop hoping it resolves — it won't until the user provides a submission_code. One question, then re-prep.

Step 2 — Walk the returned field_schema with the user

matched.field_schema.fields is the authoritative list of fields for this hackathon. Walk every field in order. For each:

  1. Show the user the field's label (not key) and description.
  2. Propose a value drawn from journey artifacts where obvious:
    • URL-typed fields (is_url: true or type: "url") for the deployed app → deployed_url from 00-state.md.
    • Project name / title field → top candidate from 01-idea.md.
    • Description / summary field → one-liner from 01-idea.md + must-haves.
    • Repo URL → ask, no default.
    • Team info → ask, no default.
    • Demo video → ask, allow skip if required: false.
    • Anything else → ask.
  3. Confirm one at a time. Do not batch. Required fields must have a non-empty value before moving on.
  4. For options-typed fields, present the allowed choices verbatim.

Use next_call.arguments.data from the prep response as the literal template — it has one key per field with a placeholder string. Replace each placeholder with the confirmed value.

Step 3 — Submit

Show the assembled data object back to the user verbatim and ask: "Submit now? (yes/no)". On yes:

{
  "tool": "prep_and_submit_hackathon_entry",
  "arguments": {
    "action": "submit",
    "hackathon_slug": "<matched.slug from prep — REQUIRED when multiple are open>",
    "app_id": "<app_id from 00-state.md — strongly recommended, unlocks +50 scoring points>",
    "submission_code": "<same code passed to prep, if any — required on first submission>",
    "data": { /* user-confirmed values keyed by field.key */ }
  }
}

Always pass hackathon_slug = matched.slug from prep. Without it, submit re-resolves and may target a different hackathon if more than one is open. Always pass app_id — scoring awards up to 50 points for Butterbase usage on that specific app.

Step 4 — Receipt

Capture the response (submission.id, submission.version, submission.updated_at, participant_created) and write docs/butterbase/05-submission.md:

# Submission

- submitted_at: <submission.updated_at>
- submission_id: <submission.id>
- version: <submission.version>
- hackathon_slug: <submission.hackathon_slug>
- hackathon_name: <matched.name>
- app_id: <submission.app_id>
- participant_created: <true|false>

## Fields submitted
<one bullet per field: label → value>

Tick - [x] submit in 00-state.md, set current_stage: done. Print a one-line success to the user including the hackathon name and submission id.

Outputs

  • Hackathon submission via prep_and_submit_hackathon_entry (action: "prep" then "submit").
  • docs/butterbase/05-submission.md.

Anti-patterns

  • ❌ Skipping action: "prep" and calling submit directly with a hardcoded field list.
  • ❌ Assuming there is exactly one open hackathon. Always check open_hackathons on the prep response.
  • ❌ Picking a hackathon for the user when prep returns multiple — always ask.
  • ❌ Inventing field keys or labels — only use what's in matched.field_schema.fields.
  • ❌ Submitting without hackathon_slug from prep (can re-resolve to the wrong hackathon).
  • ❌ Submitting without app_id when one exists (forfeits up to 50 scoring points).
  • ❌ Calling submit without explicit yes from the user on the assembled payload.

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.

Reviews can't be read right now.

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.