What you can do with Human Edge Agent Skills

5 use cases

• Turn a messy voice dump into an executable to-do list

  Paste a spoken brain dump and get actions, decisions, dependencies, and due dates in one pass.

  • Human Edge Agent Skills
  • Claude Code
  • Work and operations
  • Intermediate
  View prompt

  Using the voice-dump-to-todo Skill from FAIRY123456789/human-edge-agent-skills, convert this transcript of my spoken notes into actions, decisions, dependencies, and due dates: <transcript>. Expect a to-do list and interactive HTML. Verify that any ambiguous names are flagged rather than silently changed.

  Share Save

• Turn a spoken product idea into a coding-agent contract

  Explain a product aloud and get scope, user flow, data, constraints, and acceptance criteria.

  • Human Edge Agent Skills
  • Codex
  • Work and operations
  • Intermediate
  View prompt

  Using the vibe-to-spec Skill from the voice-to-work plugin in FAIRY123456789/human-edge-agent-skills, turn this spoken product description into a coding-agent spec: <transcript>. Expect scope, user flow, data and state, constraints, acceptance criteria, and implementation order. Verify that every acceptance criterion can be checked by a test.

  Share Save

• Test an AI trend with a 48-hour to 7-day bet

  Replace two weeks of FOMO with a small, time-boxed test that demands evidence.

  • Human Edge Agent Skills
  • Work and operations
  • Intermediate
  View prompt

  Using the trend-to-microbet Skill from FAIRY123456789/human-edge-agent-skills, design a small bet for this AI trend: <trend>, with <hours available> per week. Expect a 48-hour to 7-day test, a clear success signal, and the evidence to collect. Verify that the test can finish within the window you gave.

  Share Save

• Rewrite a correct but stiff message so it lands socially

  Keep the facts and fix the tone for an upward or formal message that is hard to answer.

  • Human Edge Agent Skills
  • Work and operations
  • Intermediate
  View prompt

  Using the social-signal-and-tact Skill from FAIRY123456789/human-edge-agent-skills, rewrite this draft message to <recipient role> so it stays factually identical but reads warmly: <draft>. Expect one revised message and a short note on what changed. Verify that no familiarity is implied that the relationship does not support.

  Share Save

• Check every Skill and plugin package before a change ships

  Run the repo's three validation scripts to catch broken Skills and stale plugin copies.

  • Human Edge Agent Skills
  • Work and operations
  • Advanced
  View prompt

  In the human-edge-agent-skills repository at <local path>, run python scripts/validate_skills.py, then sync_plugin_packages.py --check, then validate_plugin_packages.py. Expect a pass or fail result for each script with the file responsible for any failure. Verify that you have not edited anything under plugins/ by hand, since packaged copies are generated.

  Share Save

How to use Human Edge Agent Skills

1. Check you have what it needs: GitHub CLI 2.90 or newer for the gh skill commands; Codex CLI for the plugin marketplace route; Python with pyyaml for running the validation scripts.
2. Open FAIRY123456789/human-edge-agent-skills on GitHub and read the README and license.
3. Install GitHub CLI 2.90 or newer.
4. gh skill preview FAIRY123456789/human-edge-agent-skills voice-dump-to-todo (inferred, check the README)
5. gh skill install FAIRY123456789/human-edge-agent-skills voice-dump-to-todo --agent claude-code --scope user (inferred, check the README)
6. Paste the transcript of your voice notes into the agent and ask for actions, decisions, dependencies, and due dates.

These steps come from our top use case, Turn a messy voice dump into an executable to-do list, which also has a copy-paste agent prompt.

Human Edge Agent Skills alternatives · Best AI repos for Work and operations

About this repo

• Who it is forSolo builders, founders, and writers who use Claude Code, Codex, or Cursor and want reusable judgment workflows for messy voice input, specs, and decisions.
• Why it is hot
• LanguagePython
• Sourcegithub.com/FAIRY123456789/human-edge-agent-skills · Back to Hot repos

Requirements

• GitHub CLI 2.90 or newer for the gh skill commands
• Codex CLI for the plugin marketplace route
• Python with pyyaml for running the validation scripts

Related repos

• Trended earlier Spynel Spynel is a lightweight terminal program with no AI inside that gives you one chat assistant and uses your existing coding harnesses such as Codex and Claude Code to do the work.
• Trended earlier Agents Office Agents Office is a local, 3D isometric office of 35 AI agents in six departments that do real work on your own Claude login, reading and filing notes in a folder you own.
• Trended earlier Knowledge Work Plugins A collection of open source Claude plugins, each bundling skills, connectors, slash commands, and sub-agents for a knowledge worker role such as sales, legal, finance, or support.