Skip to content
Back to search

killmyidea

Describe your startup idea. Jev decides: kill it, fix it or ship it.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/monteduro--killmyidea/badge.svg)](/listings/monteduro--killmyidea)

HTML

<a href="/listings/monteduro--killmyidea"><img src="/api/listings/monteduro--killmyidea/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

npm install

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 245 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Kill My Idea
  2. Describe a startup idea. Jev decides: **KILL IT**, **FIX IT** or **SHIP IT**.
  3. No generative LLM. One request asks [TypeSafe](https://typesafe.ai) Jev 10
  4. questions in parallel: 8 plain indie-hacker questions scored 0–4, plus the
  5. category and whether the idea is understandable.
  6. ```
  7. idea → Jev (8 scores) → each × 25 → weighted average → clarity gate → KILL / FIX / SHIP
  8. ```
  9. ## 1. Install
  10. ```bash
  11. npm install
  12. ```
  13. ## 2. Environment
  14. ```bash
  15. cp .env.example .env
  16. ```
  17. ```
  18. TYPESAFE_API_KEY=your_key
  19. ```
  20. The key is only read by `api/evaluate.ts` and never reaches the browser.
  21. Successful evaluations are archived in SQLite at `data/analytics.sqlite` by
  22. default. Set `ANALYTICS_DB_PATH` to an absolute path on a persistent disk in
  23. production. Users can opt out per request with the checkbox below private history.
  24. `SITE_URL` (optional) sets the absolute URL used in the canonical and OG/Twitter
  25. tags, `robots.txt`, `sitemap.xml`, `/mcp.md` and `llms.txt` (generated at build time).
  26. It defaults to `https://killmyidea.biz`. The image itself is `public/og.png` (1200×630).
  27. For UI work without a key, set `TYPESAFE_MOCK=1`. The API then returns
  28. deterministic fake answers labelled `MOCK DATA` in the raw-data panel. Mock mode is
  29. ignored in production.
  30. ## 3. Run locally
  31. ```bash
  32. npm run dev # http://127.0.0.1:5317 (serves /api/* and /mcp/ too)
  33. npm test # unit tests
  34. npm run benchmark # balanced scoring benchmark; requires the app to be running
  35. ```
  36. Every result has a collapsed "How Jev decided" panel with raw Jev answers, probabilities, confidence,
  37. the average, the verdict, latency and token usage. `?debug=1` opens it by default.
  38. ## 4. Build
  39. ```bash
  40. npm run build # typecheck + static build in dist/
  41. ```
  42. ## 5. Deploy (Laravel Forge)
  43. The site runs on Forge as a Node server: `npm run build` for the static
  44. assets in `dist/`, then `npm run build:server && npm start` to serve the API
  45. endpoints. Set `TYPESAFE_API_KEY` in the site's environment file.
  46. SQLite needs a persistent local disk, which the Forge server provides
  47. directly (`data/analytics.sqlite`); set `ANALYTICS_DB_PATH` to an absolute
  48. path on that disk in production.
  49. Because the SPA is served with an nginx fallback (`try_files ... /index.html`),
  50. every unknown path returns the index page with HTTP 200. Bots probing random
  51. URLs therefore show up as pageviews in DataFast. To stop that, change the
  52. fallback in the site's nginx configuration (Forge → site → Edit Nginx
  53. Configuration) to return 404 for unknown paths if the app has no real routes:
  54. ```nginx
  55. location / {
  56. try_files $uri $uri/ =404;
  57. }
  58. ```
  59. ### Domain move (killmyidea.stemonte.io → killmyidea.biz)
  60. Both domains belong to the same Forge site. The old one answers every request
  61. with a 301 to the same path and query on `killmyidea.biz`, via
  62. `deploy/nginx-old-domain-redirect.conf`, pasted into the site's nginx
  63. configuration inside the `server` block. Keep that redirect for at least a year.
  64. Idea history lives in `localStorage`, which is per origin. The 301 adds
  65. `#moved`; the inline script in `index.html` then makes one hop to
  66. `public/history-transfer.html` on the old domain, which returns with its
  67. history in the URL fragment (never sent to a server) to be merged. It runs
  68. at most once per browser. Remove both once the old domain is retired.
  69. ## MCP server
  70. `api/mcp.ts` is a free, public MCP server at `/api/mcp` (Streamable HTTP, no auth).
  71. It is stateless: every POST is one JSON-RPC message answered with plain JSON, there
  72. are no sessions and GET returns 405. Tools:
  73. - `judge_idea(idea, goal?, do_not_archive?)`: same pipeline as `/api/evaluate`
  74. (`api/_judge.ts`), returns the verdict, score, best signal, biggest risk and the
  75. eight sub-scores, without the raw Jev response
  76. - `get_stats()`: analyzed count and verdict split
  77. MCP evaluations are archived like web ones, with `source = 'mcp'`.
  78. Usage is tracked in the `mcp_requests` table: one row per JSON-RPC message with the
  79. method, tool name, client name and version (sent with `initialize`), user agent,
  80. protocol version, success or error, and latency. It never stores tool arguments or
  81. IP addresses. Summary for the last 7 days (or `-- 30` for 30):
  82. ```bash
  83. npm run mcp:usage
  84. ```
  85. That report includes user agents and error messages, so it stays private. The public
  86. `/stats/` page (`stats/index.html` + `src/stats-page.tsx`, linked in the footer) reads
  87. `GET /api/usage` (`api/usage.ts`, cached 5 minutes): ideas analyzed on the web and via
  88. MCP, a 30-day daily chart, tool calls, clients and verdicts. Client and tool names come
  89. from callers, so only known ones are published (`clientLabel`), everything else is "Other".
  90. Docs: `mcp/index.html` is the static SEO page at `/mcp/` (a second Vite entry);
  91. `mcp/mcp.md` is served as `/mcp.md` for agents and `llms.txt` is generated in
  92. `vite.config.ts`. Both are emitted at build time and served by the dev server.
  93. ```bash
  94. claude mcp add --transport http killmyidea https://killmyidea.biz/api/mcp
  95. ```
  96. nginx already proxies `/api/` to the Node server, so no extra config is needed.
  97. ## 6. Scoring
  98. `src/lib/scoring.ts` and `src/lib/verdict.ts`. That's all there is:
  99. - each question: Jev score 0–4 × 25 → 0–100
  100. - **score = weighted average of the 8 questions** (`WEIGHTS`)
  101. - **Make money** weights Real problem and Money twice; **Open source** replaces Money with
  102. Adoption; **Just for fun** replaces Real problem and Money with Immediate appeal and Fun.
  103. The two goal-defining questions always count twice. The API takes `goal`
  104. (`money` default, `open_source`, `fun`).
  105. - **KILL** below 50 · **FIX** 50–64 · **SHIP** 65+ (`VERDICT_THRESHOLDS`)
  106. - if understandability is below 0.3, the raw scores are retained for analysis but the UI
  107. asks for more detail instead of presenting a score or verdict
  108. - `SCORING_VERSION` identifies the exact questions, weights, thresholds and gating semantics
  109. BEST SIGNAL / BIGGEST RISK are the highest / lowest question, with fixed copy in
  110. `src/lib/copy.ts`.
  111. ## 7. Jev questions
  112. `src/lib/questions.ts`: Real problem, Clear customer, Demand, Money, Reach,
  113. Different, Buildable, Shareable, plus Adoption, Immediate appeal and Fun for goal-specific scoring. Each has five levels written as concrete situations
  114. (Jev judges each level on its own). The HTTP client is `src/lib/typesafe.ts`.
  115. ## Layout
  116. ```
  117. api/evaluate.ts API endpoint for the website
  118. api/mcp.ts MCP server (JSON-RPC over POST /api/mcp)
  119. api/_judge.ts shared pipeline (validate → Jev → score → archive)
  120. api/stats.ts API endpoint (ideas submitted + verdict split, 5-min cache)
  121. api/usage.ts public usage numbers for /stats/ (5-min cache)
  122. api/_analytics-db.ts SQLite archive for evaluation requests and scores
  123. api/_mock.ts local-only fake answers
  124. src/lib/questions.ts Jev questions
  125. src/lib/scoring.ts 0-4 → 0-100, average
  126. src/lib/verdict.ts KILL / FIX / SHIP thresholds
  127. src/lib/copy.ts fixed copy per question
  128. src/lib/storage.ts history (IdeaStore interface, localStorage implementation)
  129. src/lib/share.ts share text and X intent URL
  130. src/lib/features.ts public site and repository URLs
  131. src/components/ UI
  132. mcp/index.html /mcp/ docs page (SEO)
  133. mcp/mcp.md /mcp.md docs for agents
  134. stats/index.html /stats/ public usage page
  135. benchmarks/cases.json balanced, synthetic benchmark cases
  136. scripts/run-scoring-benchmark.mjs benchmark runner and summary
  137. ```
  138. ## Analytics (DataFast, cookieless)
  139. Production builds load DataFast's cookieless script (`script.cookieless.js`) with
  140. the site id `dfid_UvVlC3fid8kOr95YTMIho` (`vite.config.ts`). Dev builds only track
  141. when `DATAFAST_WEBSITE_ID` is set; `DATAFAST_WEBSITE_ID=off` disables tracking.
  142. `DATAFAST_DOMAIN` defaults to `killmyidea.biz`; `DATAFAST_ALLOWED_HOSTNAMES` defaults to
  143. `killmyidea.stemonte.io` so the old domain still counts during the move. In DataFast, enable **Settings → General → Cookieless / privacy mode**
  144. so the dashboard matches the script.
  145. There is one custom goal (`src/lib/analytics.ts`): `idea_submitted`, with
  146. `length`, `saved`, and `archived` params. The idea text is never sent to DataFast.
  147. ## Evaluation archive (SQLite)
  148. Unless the user checks the opt-out, each successful evaluation creates one row
  149. in the `evaluations` table. It includes the idea, where it came from (`source`: `web` or `mcp`), scoring version, total score,
  150. raw verdict, clarity-gate status, category, all eight dimension scores, understandability, latency, model, token usage and
  151. the complete diagnostic response JSON. It deliberately excludes IP addresses,
  152. cookies and browser identifiers.
  153. For example:
  154. ```bash
  155. sqlite3 data/analytics.sqlite \
  156. 'SELECT scoring_version, goal, COUNT(*), ROUND(AVG(score), 1) FROM evaluations GROUP BY scoring_version, goal;'
  157. ```
  158. Rows collected before scoring v2 are migrated without being reinterpreted: they remain
  159. `scoring_version = 1`, while v2 stores the new `appeal` score for fun ideas. To export:
  160. ```bash
  161. sqlite3 -header -csv data/analytics.sqlite 'SELECT * FROM evaluations;' > evaluations.csv
  162. ```
  163. ## Scoring benchmark
  164. `benchmarks/cases.json` contains two similarly detailed, synthetic ideas for each category.
  165. The runner evaluates every case against all three goals and always sends `doNotArchive: true`.
  166. With the app running against Jev:
  167. ```bash
  168. BENCHMARK_RUNS=3 npm run benchmark
  169. ```
  170. It reports score and verdict distributions by goal, category accuracy, clarity gates and
  171. repeat-score range. The full JSON is written to the ignored
  172. `benchmarks/results/latest.json`; set `BENCHMARK_URL` to test a deployed instance.
  173. `.mcp.json` registers the official DataFast MCP server (`https://datafa.st/api/mcp`).
  174. In Claude Code, run `/mcp` and sign in with OAuth to query these analytics.
  175. ## Privacy
  176. The public policy is `privacy/index.html` (`/privacy/`), with the terms in `terms/index.html` (`/terms/`).
  177. Keep them in step with what the code stores. Contact: `hello@killmyidea.biz` (Cloudflare Email Routing).
  178. - Private history is off by default and stays in the browser's `localStorage`. There are no accounts.
  179. - Successful evaluations are archived server-side for product analysis by default; the form provides an explicit opt-out.
  180. - DataFast never receives idea text. To move private history to Supabase, implement
  181. `IdeaStore` in `src/lib/storage.ts`.
  182. ## License
  183. MIT. See [LICENSE](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.