cursi.dev Indie Product Toolkit
A practical toolkit for solo developers and small teams shipping reliable products. Turn client feedback into build-ready work, transfer standards between repositories, isolate Supabase roles, audit financial data provenance, and run a final iOS release gate. Five focused skills cover the path from incoming requests to a safer, better-evidenced release.
Works with any agent that signs in to Agensi, such as Claude, ChatGPT, Claude Code, Cursor and Codex CLI.
One payment, lifetime access. 5 skills unlock instantly in your library.
What's included
5 skills
Turn client messages and screenshots into source-linked requirements, testable tickets, and an agent handoff.
Audits financial app data for traceable sources, stale records, duplicates, and normalization errors.
Evidence-based iOS release audit with a PASS, BLOCK, or UNKNOWN decision and prioritized fixes.
Adapt engineering practices from a reference repo to a target with an adopt/adapt/skip/blocked matrix.
Maps app roles to Supabase RLS policies and creates cross-user and cross-tenant access tests.
See it in action
Client Feedback to Build Brief
Example session with this skill installed
Turn these notes into a build brief and agent prompt. Do not code or create issues.
S01 — Client owner, 09:00: “Change the submit label to ‘Send request’. Keep email/password login.”
S02 — Same owner, 09:10: “After submission, show a receipt number. If it fails, preserve everything I typed.”
S03 — Designer, 09:15: “Maybe remove login completely and use a public link?”
S04 — Client owner, 09:20: “For this release, keep login. Public access can wait.”
No repository, receipt format, or backend contract is provided.
- Read your context and instructions
- Compiled the client feedback to
Scope: improve the signed-in submission flow. Preserve email/password login. Public access is deferred by S04. No code or external issue was created.
| ID | Requirement | Sources | Status |
|---|---|---|---|
| R01 | Submit control reads exactly “Send request” | S01 | READY |
| R02 | Success displays a receipt number | S02 | NEEDS DECISION: source/format |
| R03 | Failed submission preserves entered values | S02 | READY; current behavior unverified |
| R04 | Keep email/password login | S01, S04 | READY constraint |
| R05 | Public access | S03, S04 | DEFERRED |
Decision D01: What generates the receipt and what is its display format? Do not fabricate an endpoint, database field, or number.
T01 — Update submit copy. Sources S01; requirements R01/R04. Acceptance: the signed-in submission control reads exactly “Send request”; login remains required. Discover the control and localization pattern. No file path is known.
T02 — Preserve values after failure. Source S02; requirements R03/R04. Acceptance: given a completed form and a failed submit request, entered values remain available for retry and success is not shown. Proposed extra check for confirmation: prevent duplicate creation on retry.
T03 — Show receipt on confirmed success. Source S02; R02. Blocked by D01 and discovery of the response contract. Acceptance after D01: confirmed success displays its agreed receipt; failure never displays a fabricated receipt.
Suggested product priority: prevent lost input first, receipt confirmation second, copy third. Implementation may begin with T01 independently; T03 stays blocked. Effort, owners, dates, and validation commands are unknown.
Agent handoff: inspect repository instructions, login boundary, submit control, localization, and request state. Implement T01/T02 only if coding is authorized. Preserve login. Hold T03 until D01 is answered. Discover relevant checks and report actual results. Do not deploy or create issues without a separate request.
Coverage: all five requirements map to tickets, a protected constraint, or a deferred item. This is a fictional worked example, not a production run.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads these skills and uses them on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every skill in this bundle goes through our trust checks.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.