Skip to content
Back to search

lateral-thinking

Lateral thinking skills for AI agents - 8 techniques + a router. de Bono for your coding agent.

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/danium--lateral-thinking/badge.svg)](/listings/danium--lateral-thinking)

HTML

<a href="/listings/danium--lateral-thinking"><img src="/api/listings/danium--lateral-thinking/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.

**Any agent (recommended):**

npx skills add danium/lateral-thinking

**Claude Code (plugin):**

/plugin marketplace add danium/lateral-thinking
/plugin install lateral-thinking@lateral-thinking

**Codex:** ask `$skill-installer` to install from `https://github.com/danium/lateral-thinking`

**Manual:** clone and copy `skills/*` into your agent's skills directory.

> The toolkit installs as one bundle — the `lateral` router needs its sibling techniques.

---

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.

.claude-plugin/marketplace.json · Manifest · 11 lines

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

  1. {
  2. "name": "lateral-thinking",
  3. "owner": { "name": "danium" },
  4. "plugins": [
  5. {
  6. "name": "lateral-thinking",
  7. "source": ".",
  8. "description": "Eight lateral thinking techniques plus a router that picks the right one for how you're stuck."
  9. }
  10. ]
  11. }
README.md · README · 151 lines

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

  1. # lateral-thinking
  2. **Your product isn't out of ideas. It's stuck in the obvious ones.**
  3. [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
  4. ![Skills: 10](https://img.shields.io/badge/skills-10-blue.svg)
  5. Give your AI coding agent eight structured ways to find a better product direction — not just another feature to ship. A router diagnoses how you're stuck and picks the right move.
  6. ![A maze breaking into unexpected paths, captioned: Your next feature isn't the next obvious one.](assets/social-preview.jpg)
  7. ---
  8. ## The demo
  9. Same agent. Same prompt. One of them read a skill first.
  10. > **Prompt:** *"Our users create one project, then never return. What product direction would make that first project worth coming back to?"*
  11. <table>
  12. <tr>
  13. <th width="50%">Any agent, vanilla</th>
  14. <th width="50%">Same agent, <code>random-stimulus</code></th>
  15. </tr>
  16. <tr valign="top">
  17. <td>
  18. **Onboarding checklist**
  19. Give them three next steps after project creation.
  20. **Weekly email summary**
  21. Remind them what changed and invite them back.
  22. **Personalized dashboard**
  23. Make the home screen more useful after the first project.
  24. *(Three sensible features. One idea: ask the user to remember the product.)*
  25. </td>
  26. <td>
  27. 🗼 **The lighthouse beam**
  28. It does not light the whole sea. It sends a precise signal when navigation matters.
  29. → A return trigger should be caused by the user's project, not by a generic reminder.
  30. **Beacon** — when a project reaches a date, risk, or decision the user chose at creation, send one concise signal explaining why it needs attention now.
  31. 🍞 **The sourdough starter**
  32. It becomes more useful between visits. Small inputs create a living thing worth checking on.
  33. → A first project should begin a rhythm, not become a static container.
  34. **Pulse** — turn the first project into a lightweight weekly brief: what changed, what is blocked, and the one decision that would move it forward.
  35. 🎡 **The Ferris wheel** — it repeats a visible loop, but the force-fit only produced a prettier recurring dashboard. **Abandoned.**
  36. </td>
  37. </tr>
  38. </table>
  39. The vanilla answer isn't bad. It's *good* — and every feature asks the user to do more work or remember to return. Three features, one idea.
  40. The technique produced different **product directions**: a project that signals at the right moment, and a project that gets more useful between visits. The meta-pattern is the point: retention is not a reminder problem. It is a **new-signal problem**. Give people a reason to return that did not exist when they left.
  41. ---
  42. ## Install
  43. **Any agent (recommended):**
  44. ```bash
  45. npx skills add danium/lateral-thinking
  46. ```
  47. **Claude Code (plugin):**
  48. ```
  49. /plugin marketplace add danium/lateral-thinking
  50. /plugin install lateral-thinking@lateral-thinking
  51. ```
  52. **Codex:** ask `$skill-installer` to install from `https://github.com/danium/lateral-thinking`
  53. **Manual:** clone and copy `skills/*` into your agent's skills directory.
  54. > The toolkit installs as one bundle — the `lateral` router needs its sibling techniques.
  55. ---
  56. ## Pick your technique
  57. | Symptom | Technique |
  58. |---|---|
  59. | Ideas all feel the same; brainstorm output is predictable | [`random-stimulus`](skills/random-stimulus/SKILL.md) |
  60. | A constraint or rule feels unbreakable | [`provocation`](skills/provocation/SKILL.md) |
  61. | Requirements assume things nobody has questioned | [`inversion`](skills/inversion/SKILL.md) |
  62. | We might be solving the wrong problem | [`concept-fan`](skills/concept-fan/SKILL.md) |
  63. | The solution works but feels derivative | [`analogy`](skills/analogy/SKILL.md) |
  64. | We have one idea and need variations | [`scamper`](skills/scamper/SKILL.md) |
  65. | A decision is being made too fast / everyone agrees | [`six-hats`](skills/six-hats/SKILL.md) |
  66. | Everything feels timid, safe, cautious | [`worst-idea`](skills/worst-idea/SKILL.md) |
  67. **Applied:** [`post`](skills/post/SKILL.md) — a post on a topic or a reply to someone's post, routed through the toolkit and shaped into drafts you can send.
  68. Or just tell your agent **"use the lateral skill"** and let it diagnose you.
  69. ---
  70. ## Anatomy of a session
  71. Every technique in this repo makes failures visible. Here is `scamper` refusing to pad its own output, mid-session:
  72. > **Substitute:** nothing new. The command is already swapped down to one line, and the only meaningful substitution (git for installer) already ships as a channel. Moving on.
  73. >
  74. > **Put to other use:** empty. The install does one job and generates no by-product worth repointing. Moving on.
  75. >
  76. > **Reverse / Rearrange:** empty as a distinct idea. The one real reversal (router-first, siblings later) is Variant D under another name, so it counts once. Nothing else to flip. Moving on.
  77. Three of seven operations came up empty — and the last one had a genuine hit that it *disqualified for being a duplicate.*
  78. This is the whole point. An agent that produces seven strong ideas from seven prompts is pattern-matching on the shape of a good answer, and you can't tell which ideas it actually believes. Every skill here carries an **abandonment rule** — a stated condition for giving up, and an instruction to show you when it does. The visible failures are what make the rest credible.
  79. The second requirement is a **meta-pattern step**: after the batch, scan across what landed and name the structural insight. In that `scamper` session, the empty operations *were* the finding — the install command's surface is already tight, so the softness had to be somewhere else. It was: in the unstated dependency between the router and its siblings.
  80. ---
  81. ## What this is (and isn't)
  82. These are real methods, sourced:
  83. - **Random Stimulus**, **Provocation (Po)**, **Concept Fan** — Edward de Bono, *Lateral Thinking* (1970), *Po: Beyond Yes and No* (1972), *Serious Creativity* (1992)
  84. - **Six Thinking Hats®** — Edward de Bono (1985). A registered trademark of the de Bono Group; this is an independent educational implementation.
  85. - **SCAMPER** — Bob Eberle (1971), building on Alex Osborn's checklist
  86. - **Forced Analogy** — Synectics, William J.J. Gordon (1961)
  87. - **Assumption Inversion** — classical dialectic; "invert, always invert" (Carl Jacobi, via Charlie Munger)
  88. - **Worst Possible Idea** — reverse brainstorming; standard in the d.school/IDEO toolkit
  89. **This is not magic and it is not a prompt pack.** It is structured divergence with quality rules attached. The techniques constrain *how* an agent generates, and the honesty mechanics constrain what it's allowed to claim afterward.
  90. **On invocation:** auto-invocation is best-effort — your agent may or may not reach for a skill on its own, and that depends on the agent. For the reliable path, name the skill or ask your agent to use the router.
  91. ---
  92. ## Contributing
  93. New techniques need provenance and honesty mechanics. Worked examples must come from real sessions. See [CONTRIBUTING.md](CONTRIBUTING.md).
  94. ## License
  95. [MIT](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

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