Skip to content
Back to search

envisioning

MarketplaceWorks with Claude CodeGitHub

Agent skills and plugins from Envisioning.

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/envisioning--foresight-skill/badge.svg)](/listings/envisioning--foresight-skill)

HTML

<a href="/listings/envisioning--foresight-skill"><img src="/api/listings/envisioning--foresight-skill/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

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

Claude Code

In a Claude Code session, run:

/plugin marketplace add envisioning/foresight-skill
/plugin install envisioning-foresight@envisioning

The plugin adds both skills and connects the envisioning.com MCP server.

- Ask for a horizon scan, a technology radar or "what is next for" a field, and Claude uses the foresight skill. Call it directly with `/envisioning-foresight:foresight`. - Ask "is this forecast credible?", "will X really happen by 2030?" or "how did past predictions about X turn out?", or paste a headline about the future, and Claude uses the forecast check. Call it directly with `/envisioning-foresight:forecast-check`, followed by the claim.

Claude.ai and the Claude desktop app

1. Download this repository and zip the `skills/foresight` folder, the `skills/forecast-check` folder, or both as separate zips. 2. In Claude, open Settings, then Capabilities, and upload each zip under Skills. 3. To give Claude the data tools too, add a custom connector with the URL `https://www.envisioning.com/mcp`.

Other agents

Both skills follow the Agent Skills format, so they work in any agent that supports that format. Copy the folders under `skills/` to wherever your agent loads skills from. Add `https://www.envisioning.com/mcp` as a remote MCP server (Streamable HTTP). If your agent has no MCP support, each skill tells it how to read the same data from https://www.envisioning.com/llms.txt and from Markdown versions of each page.

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 · 22 lines
  1. {
  2. "name": "envisioning",
  3. "description": "Agent skills and plugins from Envisioning.",
  4. "owner": {
  5. "name": "Envisioning",
  6. "url": "https://www.envisioning.com"
  7. },
  8. "plugins": [
  9. {
  10. "name": "envisioning-foresight",
  11. "source": "./",
  12. "description": "Run Envisioning's foresight method and check forecasts against graded past forecasts, citing envisioning.com research.",
  13. "category": "research",
  14. "tags": [
  15. "foresight",
  16. "horizon scanning",
  17. "radar",
  18. "forecasting"
  19. ]
  20. }
  21. ]
  22. }
README.md · README · 59 lines
  1. # Envisioning foresight skills
  2. These two skills let your AI agent use Envisioning's foresight method and research.
  3. - **foresight** runs the full method. The agent frames a strategic question, scans for signals of change, checks each claim against its sources, places the signals by category, readiness and horizon, checks timing claims against graded past forecasts, and writes a radar with every source cited.
  4. - **forecast-check** checks one claim about the future. The agent pins the claim down (what, by whom, by when, measured how), looks up how similar published forecasts were graded, reads the readiness of the technology involved, and tells you how much weight to give the claim, with every source cited.
  5. The agent pulls data from Envisioning's open research: thousands of technologies with readiness scores, a technology vocabulary, and Hindsight, a dataset of published forecasts graded against what happened. It cites envisioning.com permalinks for everything it uses from us.
  6. The skills are free and public.
  7. ## What is in the repository
  8. - `skills/foresight/SKILL.md`: the method as steps an agent follows.
  9. - `skills/foresight/references/`: the full method with its sources, data sources and citation rules, scan types, the Hindsight check, the output format, and short definitions.
  10. - `skills/forecast-check/SKILL.md`: the forecast check as steps an agent follows.
  11. - `.claude-plugin/`: the Claude Code plugin and marketplace manifests.
  12. - `.mcp.json`: the envisioning.com MCP server (`https://www.envisioning.com/mcp`, read-only, no sign-in).
  13. ## Install
  14. ### Claude Code
  15. In a Claude Code session, run:
  16. ```
  17. /plugin marketplace add envisioning/foresight-skill
  18. /plugin install envisioning-foresight@envisioning
  19. ```
  20. The plugin adds both skills and connects the envisioning.com MCP server.
  21. - Ask for a horizon scan, a technology radar or "what is next for" a field, and Claude uses the foresight skill. Call it directly with `/envisioning-foresight:foresight`.
  22. - Ask "is this forecast credible?", "will X really happen by 2030?" or "how did past predictions about X turn out?", or paste a headline about the future, and Claude uses the forecast check. Call it directly with `/envisioning-foresight:forecast-check`, followed by the claim.
  23. ### Claude.ai and the Claude desktop app
  24. 1. Download this repository and zip the `skills/foresight` folder, the `skills/forecast-check` folder, or both as separate zips.
  25. 2. In Claude, open Settings, then Capabilities, and upload each zip under Skills.
  26. 3. To give Claude the data tools too, add a custom connector with the URL `https://www.envisioning.com/mcp`.
  27. ### Other agents
  28. Both skills follow the Agent Skills format, so they work in any agent that supports that format. Copy the folders under `skills/` to wherever your agent loads skills from. Add `https://www.envisioning.com/mcp` as a remote MCP server (Streamable HTTP). If your agent has no MCP support, each skill tells it how to read the same data from https://www.envisioning.com/llms.txt and from Markdown versions of each page.
  29. ## Data the skill uses
  30. - envisioning.com research hubs, technology library and vocabulary
  31. - Hindsight: https://www.envisioning.com/hindsight and https://github.com/envisioning/hindsight
  32. - The National Capability Benchmark: https://ncb.envisioning.com
  33. - The methodology chapters: https://www.envisioning.com/methodology
  34. To run scans continuously across many questions and models, Envisioning Signals runs the same method as a platform: https://signals.envisioning.com
  35. ## License
  36. The skill text is CC BY 4.0. The code and configuration are MIT. See `NOTICE.md` and `LICENSE`.
  37. Made by Envisioning, https://www.envisioning.com

Ratings and comments

Sign in to rate this skill.

Sign in

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