humanmd 0.1.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- humanmd-0.1.0/.github/ISSUE_TEMPLATE/documentation.md +13 -0
- humanmd-0.1.0/.github/ISSUE_TEMPLATE/example.md +16 -0
- humanmd-0.1.0/.github/ISSUE_TEMPLATE/feature.md +16 -0
- humanmd-0.1.0/.github/pull_request_template.md +6 -0
- humanmd-0.1.0/.github/workflows/tests.yml +36 -0
- humanmd-0.1.0/.gitignore +8 -0
- humanmd-0.1.0/AGENTS.md +11 -0
- humanmd-0.1.0/CONTRIBUTING.md +22 -0
- humanmd-0.1.0/HUMAN.md +19 -0
- humanmd-0.1.0/LICENSE +21 -0
- humanmd-0.1.0/PKG-INFO +218 -0
- humanmd-0.1.0/README.md +203 -0
- humanmd-0.1.0/docs/agent-integration.md +13 -0
- humanmd-0.1.0/docs/assets/humanmd-banner.framing.prompt.txt +3 -0
- humanmd-0.1.0/docs/assets/humanmd-banner.png +0 -0
- humanmd-0.1.0/docs/assets/humanmd-banner.prompt.txt +4 -0
- humanmd-0.1.0/docs/assets/humanmd-banner.title-only.prompt.txt +3 -0
- humanmd-0.1.0/docs/assets/humanmd-footer.capitalization.prompt.txt +3 -0
- humanmd-0.1.0/docs/assets/humanmd-footer.png +0 -0
- humanmd-0.1.0/docs/assets/humanmd-footer.prompt.txt +5 -0
- humanmd-0.1.0/docs/assets/think-before-code.framing.prompt.txt +3 -0
- humanmd-0.1.0/docs/assets/think-before-code.loop.prompt.txt +3 -0
- humanmd-0.1.0/docs/assets/think-before-code.png +0 -0
- humanmd-0.1.0/docs/assets/think-before-code.prompt.txt +3 -0
- humanmd-0.1.0/docs/assets/think-before-code.researchers.prompt.txt +3 -0
- humanmd-0.1.0/docs/cli.md +85 -0
- humanmd-0.1.0/docs/philosophy.md +15 -0
- humanmd-0.1.0/docs/release-notes-0.1.0.md +21 -0
- humanmd-0.1.0/docs/releasing.md +61 -0
- humanmd-0.1.0/docs/specification.md +33 -0
- humanmd-0.1.0/examples/research-project/HUMAN.md +34 -0
- humanmd-0.1.0/examples/software-project/HUMAN.md +31 -0
- humanmd-0.1.0/hatch_build.py +17 -0
- humanmd-0.1.0/pyproject.toml +40 -0
- humanmd-0.1.0/src/humanmd/__init__.py +1 -0
- humanmd-0.1.0/src/humanmd/__main__.py +3 -0
- humanmd-0.1.0/src/humanmd/cli.py +149 -0
- humanmd-0.1.0/src/humanmd/templates/exploration.md +34 -0
- humanmd-0.1.0/src/humanmd/templates/minimal.md +23 -0
- humanmd-0.1.0/src/humanmd/templates/research.md +37 -0
- humanmd-0.1.0/src/humanmd/templates/software.md +34 -0
- humanmd-0.1.0/tests/test_cli.py +266 -0
- humanmd-0.1.0/uv.lock +979 -0
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: Documentation improvement
|
|
3
|
+
about: Help make the convention easier to understand.
|
|
4
|
+
title: ''
|
|
5
|
+
labels: ''
|
|
6
|
+
assignees: ''
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
## What was confusing?
|
|
10
|
+
Link to the affected section.
|
|
11
|
+
|
|
12
|
+
## Suggested clarification
|
|
13
|
+
What would have helped?
|
|
@@ -0,0 +1,16 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: Real-world example
|
|
3
|
+
about: Share how you use HUMAN.md.
|
|
4
|
+
title: ''
|
|
5
|
+
labels: ''
|
|
6
|
+
assignees: ''
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
## Project context
|
|
10
|
+
Share only information you have permission to publish.
|
|
11
|
+
|
|
12
|
+
## How HUMAN.md helped or fell short
|
|
13
|
+
Distinguish observations from expectations.
|
|
14
|
+
|
|
15
|
+
## Example and permission
|
|
16
|
+
Provide a link or excerpt and confirm permission to share it.
|
|
@@ -0,0 +1,16 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: Feature proposal
|
|
3
|
+
about: Suggest a small improvement that preserves human ownership.
|
|
4
|
+
title: ''
|
|
5
|
+
labels: ''
|
|
6
|
+
assignees: ''
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
## Problem and motivation
|
|
10
|
+
Who needs this, and why?
|
|
11
|
+
|
|
12
|
+
## Proposed improvement
|
|
13
|
+
How would it help without adding unnecessary complexity?
|
|
14
|
+
|
|
15
|
+
## Alternatives and effect on project intent
|
|
16
|
+
Does it change any goals or assumptions in HUMAN.md?
|
|
@@ -0,0 +1,36 @@
|
|
|
1
|
+
name: Tests
|
|
2
|
+
|
|
3
|
+
on:
|
|
4
|
+
push:
|
|
5
|
+
branches: [main]
|
|
6
|
+
pull_request:
|
|
7
|
+
workflow_dispatch:
|
|
8
|
+
|
|
9
|
+
permissions:
|
|
10
|
+
contents: read
|
|
11
|
+
|
|
12
|
+
jobs:
|
|
13
|
+
test:
|
|
14
|
+
name: ${{ matrix.os }} / Python ${{ matrix.python }}
|
|
15
|
+
runs-on: ${{ matrix.os }}
|
|
16
|
+
strategy:
|
|
17
|
+
fail-fast: false
|
|
18
|
+
matrix:
|
|
19
|
+
os: [ubuntu-latest, windows-latest, macos-latest]
|
|
20
|
+
python: ['3.10', '3.14']
|
|
21
|
+
steps:
|
|
22
|
+
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
|
|
23
|
+
- name: Install uv
|
|
24
|
+
uses: astral-sh/setup-uv@c771a70e6277c0a99b617c7a806ffedaca235ff9 # v9.0.0
|
|
25
|
+
with:
|
|
26
|
+
version: '0.9.8'
|
|
27
|
+
- name: Install locked development environment
|
|
28
|
+
run: uv sync --locked --python ${{ matrix.python }}
|
|
29
|
+
- name: Test CLI and installed distributions
|
|
30
|
+
run: uv run --locked pytest -ra
|
|
31
|
+
- name: Build release distributions
|
|
32
|
+
run: uv build
|
|
33
|
+
- name: Validate package metadata
|
|
34
|
+
run: uv run --locked twine check dist/*
|
|
35
|
+
- name: Render README
|
|
36
|
+
run: uv run --locked python -m readme_renderer README.md --output "${{ runner.temp }}/humanmd-readme.html"
|
humanmd-0.1.0/.gitignore
ADDED
humanmd-0.1.0/AGENTS.md
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
1
|
+
# Agent guidance
|
|
2
|
+
|
|
3
|
+
- Read HUMAN.md before implementation and preserve its documented philosophy.
|
|
4
|
+
- Keep the CLI dependency-light and avoid unnecessary features.
|
|
5
|
+
- Ask before changing fundamental project direction.
|
|
6
|
+
- Suggest human-reviewed updates when findings change the reasoning; no formal Decision Log is required.
|
|
7
|
+
- Never silently rewrite human-owned intent.
|
|
8
|
+
- Use uv for package management: `uv sync --locked`, `uv run --locked pytest`, and `uv build`. Commit intentional dependency changes in `uv.lock`.
|
|
9
|
+
- Run pytest and verify built distributions when changing CLI or packaging behavior.
|
|
10
|
+
- Character illustrations must include both a woman and a man depicted as active coders or researchers alongside the AI robot. Give both people equal agency and prominence. Keep the modern pixel-art style and show human thinking before coding in process illustrations. Text-only wordmarks can use a simple pixel grid without characters.
|
|
11
|
+
- Keep artwork minimal and positive: simple pixel shapes, restrained colors, generous whitespace, few props, and readable text. Prefer one clear symbol per process step to repeated detailed character scenes.
|
|
@@ -0,0 +1,22 @@
|
|
|
1
|
+
# Contributing
|
|
2
|
+
|
|
3
|
+
Read [HUMAN.md](HUMAN.md) and [AGENTS.md](AGENTS.md) before changing the project. Keep adoption simple and human intent human-owned.
|
|
4
|
+
|
|
5
|
+
Use an issue for a feature proposal, documentation improvement, or real-world example. Explain the problem and why it matters. Discuss fundamental changes in direction with the maintainer before implementing them. Clearly label fictional examples and proposed experiments; do not invent results or adoption claims.
|
|
6
|
+
|
|
7
|
+
For development, use Python 3.10+ and [uv](https://docs.astral.sh/uv/getting-started/installation/):
|
|
8
|
+
|
|
9
|
+
```bash
|
|
10
|
+
uv sync --locked
|
|
11
|
+
uv run --locked pytest
|
|
12
|
+
uv build
|
|
13
|
+
uv run --locked twine check dist/*
|
|
14
|
+
```
|
|
15
|
+
|
|
16
|
+
The test-only GitHub Actions workflow covers Linux, Windows, and macOS on Python 3.10 and 3.14. It does not publish packages or change repository settings. See the [release guide](docs/releasing.md) for preparing artifacts and reviewing platform results.
|
|
17
|
+
|
|
18
|
+
uv creates `.venv` and installs the project in editable mode with its development dependency group. Commit `uv.lock` when dependencies change; use `uv add --dev PACKAGE` for development tools and `uv lock --upgrade` for intentional dependency updates. Review the resulting lockfile diff.
|
|
19
|
+
|
|
20
|
+
Add meaningful tests for CLI behavior changes and verify bundled templates when changing packaging. Keep the four canonical templates in `src/humanmd/templates/`; do not duplicate them. Check documentation links and suggest updates when meaningful findings change the reasoning; a formal Decision Log is optional.
|
|
21
|
+
|
|
22
|
+
Submit a focused pull request explaining the problem, resulting behavior, and checks performed. Avoid new runtime dependencies, automation, or unrelated features without a demonstrated need. Contributions are provided under the repository's [MIT license](LICENSE).
|
humanmd-0.1.0/HUMAN.md
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
# HUMAN.md
|
|
2
|
+
|
|
3
|
+
## A note from the creator
|
|
4
|
+
|
|
5
|
+
With the emergence of agentic coding tools, development feels much faster to me. I see real value in being able to explore an idea and turn it into something working more easily. But I also worry that the emphasis is drifting toward producing more, with too little attention to whether the work is useful, considered, or worth maintaining. I want us to use these tools to do worthwhile things well.
|
|
6
|
+
|
|
7
|
+
That concern is where HUMAN.md began. In a world that seems to move faster all the time, especially in technology, AI, and development, I want to make a little room to slow down and collect our thoughts. For me, HUMAN.md is also a small, symbolic invitation: take a pen or an iPad, make a good cup of coffee or tea, and give yourself time to think about what you want to achieve with the project.
|
|
8
|
+
|
|
9
|
+
Before writing code or asking an agent to generate it, I want to pause and ask: why should this project exist? What am I trying to achieve? What questions am I asking? What do I expect to learn? I don't need all the answers before starting. An experiment can be a good first step if I can explain what I hope to learn from it.
|
|
10
|
+
|
|
11
|
+
I can ask an agent to help explore an idea, question my assumptions, and carry out the work. I still need to choose the direction, review the results, and take responsibility for the code I use and how I use it. Having something that runs doesn't settle whether it serves the purpose I had in mind.
|
|
12
|
+
|
|
13
|
+
As we discussed this idea, I realized that focusing only on the beginning was too narrow. My questions and assumptions can change while building, and an unexpected result can change what I think the project should become. I want to leave enough of that reasoning visible that someone else, or I myself later, can understand how the direction developed.
|
|
14
|
+
|
|
15
|
+
I like to think of HUMAN.md as a fireplace for that process: a place where you and your team can gather your thoughts, talk through what matters, and return when the direction feels unclear. It doesn't have to begin in a Markdown file. The thinking might happen on paper, on a tablet, or in a conversation; HUMAN.md can give what you want to carry forward a home alongside the project.
|
|
16
|
+
|
|
17
|
+
That is the commitment I want HUMAN.md to represent. I don't have a perfect structure for these thoughts, and I don't think every project needs the same one. A note like this, open questions, reflections on findings, or links to research and decisions can all be useful. The structure should follow the project. I don't want filling in headings to take the place of thinking.
|
|
18
|
+
|
|
19
|
+
I don't know yet whether a small Markdown file will help with this. I want to try it, listen to people who question the idea, and learn from using it. This is my thinking so far, and I expect it to change.
|
humanmd-0.1.0/LICENSE
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026 HUMAN.md contributors
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
humanmd-0.1.0/PKG-INFO
ADDED
|
@@ -0,0 +1,218 @@
|
|
|
1
|
+
Metadata-Version: 2.5
|
|
2
|
+
Name: humanmd
|
|
3
|
+
Version: 0.1.0
|
|
4
|
+
Summary: Initialize human-owned research and software project intent documents from Markdown templates.
|
|
5
|
+
Project-URL: Repository, https://github.com/Kylo1110/HUMANmd
|
|
6
|
+
Project-URL: Issues, https://github.com/Kylo1110/HUMANmd/issues
|
|
7
|
+
License-Expression: MIT
|
|
8
|
+
License-File: LICENSE
|
|
9
|
+
Keywords: ai,human-intent,markdown,templates
|
|
10
|
+
Classifier: Development Status :: 3 - Alpha
|
|
11
|
+
Classifier: Programming Language :: Python :: 3
|
|
12
|
+
Classifier: Programming Language :: Python :: 3 :: Only
|
|
13
|
+
Requires-Python: >=3.10
|
|
14
|
+
Description-Content-Type: text/markdown
|
|
15
|
+
|
|
16
|
+
<p align="center">
|
|
17
|
+
<img src="https://raw.githubusercontent.com/Kylo1110/HUMANmd/main/docs/assets/humanmd-banner.png" width="100%" alt="HUMAN.md. A pixel-art woman and man, depicted as developers or researchers, collaborate with a friendly robot over a shared intent document.">
|
|
18
|
+
</p>
|
|
19
|
+
|
|
20
|
+
<p align="center">
|
|
21
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-teal.svg?style=flat-square" alt="License: MIT"></a>
|
|
22
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/pyproject.toml"><img src="https://img.shields.io/badge/python-3.10%2B-blue.svg?style=flat-square" alt="Python: 3.10+"></a>
|
|
23
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/pyproject.toml"><img src="https://img.shields.io/badge/version-0.1.0-orange.svg?style=flat-square" alt="Version: 0.1.0"></a>
|
|
24
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/docs/cli.md#errors-and-safety"><img src="https://img.shields.io/badge/works-offline-teal.svg?style=flat-square" alt="Works offline"></a>
|
|
25
|
+
</p>
|
|
26
|
+
|
|
27
|
+
<p align="center">
|
|
28
|
+
<a href="https://github.com/Kylo1110/HUMANmd#get-started">Get started</a> ·
|
|
29
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/docs/specification.md">Specification</a> ·
|
|
30
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/docs/agent-integration.md">Agent guidance</a> ·
|
|
31
|
+
<a href="https://github.com/Kylo1110/HUMANmd#contributing">Contribute</a>
|
|
32
|
+
</p>
|
|
33
|
+
|
|
34
|
+
## Code explains how. HUMAN.md explains why.
|
|
35
|
+
|
|
36
|
+
HUMAN.md is an invitation to think deliberately before building, and to keep thinking as a project develops. Ask why the work matters, what you hope to achieve, and which assumptions you want to test. A small, human-owned Markdown file gives the important parts of that reflection a home, so collaborators, AI agents, and your future self can understand how the direction developed.
|
|
37
|
+
|
|
38
|
+
A few honest sentences can be enough. Templates, headings, and Decision Logs are optional: adapt them as the project develops, leave questions open, and add detail when it helps. HUMAN.md keeps the reasoning alongside your code and existing documentation.
|
|
39
|
+
|
|
40
|
+
## A note from the creators
|
|
41
|
+
|
|
42
|
+
In a world that keeps moving faster, we want to make a little room to slow down and collect our thoughts. Take a pen or an iPad, make a cup of coffee or tea, and think about what you want your project to achieve. We hope it can become a small fireplace for you and your team—a place to gather your thoughts, question assumptions, and return as the project develops.
|
|
43
|
+
|
|
44
|
+
**There is no perfect structure to fill in. Give the thinking a place to happen.**
|
|
45
|
+
|
|
46
|
+
Agents can help us explore and challenge ideas, while we remain responsible for directing their work, reviewing the results, and how we use the code. A [recent study of repository-level context files](https://arxiv.org/html/2602.11988v3) found that they did not generally improve coding agents' task success. For us, that's a reminder to focus HUMAN.md on making human intent explicit, rather than adding documentation for its own sake.
|
|
47
|
+
|
|
48
|
+
HUMAN.md is an evolving idea. We don't claim it's a proven solution or the right fit for every project, and we invite you to help shape it. Our [creator's note](https://github.com/Kylo1110/HUMANmd/blob/main/HUMAN.md#a-note-from-the-creator) shares the personal thinking behind the idea.
|
|
49
|
+
|
|
50
|
+
## Before the first line of code
|
|
51
|
+
|
|
52
|
+
**Start with a reason for the next step.**
|
|
53
|
+
|
|
54
|
+
<p align="center">
|
|
55
|
+
<img src="https://raw.githubusercontent.com/Kylo1110/HUMANmd/main/docs/assets/think-before-code.png" width="100%" alt="A blonde woman researcher and a man researcher, wearing lab coats and holding research notes, work with an AI robot through six steps: find the purpose, question assumptions, define success, write HUMAN.md, make a human decision, then build. A return loop from the decision to the questions shows reflection and refinement before proceeding.">
|
|
56
|
+
</p>
|
|
57
|
+
|
|
58
|
+
**Your first step may be an experiment. Write down what you want to learn before asking an agent to build it.**
|
|
59
|
+
|
|
60
|
+
The diagram offers a starting point, with room to revisit questions, assumptions, and intended outcomes. An exhaustive plan isn't required. As you experiment, implement, and maintain the project, explain what you learn and why the approach changes.
|
|
61
|
+
|
|
62
|
+
HUMAN.md follows three simple principles:
|
|
63
|
+
|
|
64
|
+
1. Human ownership. People define the purpose and remain responsible for the work.
|
|
65
|
+
2. Agent awareness. Give agents context to question assumptions and ask before changing direction.
|
|
66
|
+
3. Evolving reasoning. Record meaningful changes in understanding, not every code change.
|
|
67
|
+
|
|
68
|
+
## One file. A different purpose.
|
|
69
|
+
|
|
70
|
+
Your README may already explain why the project exists. HUMAN.md offers a dedicated home when it helps to keep ideas, assumptions, and decisions together. Link to existing explanations rather than duplicating them. These roles can overlap:
|
|
71
|
+
|
|
72
|
+
| File | Purpose |
|
|
73
|
+
| --- | --- |
|
|
74
|
+
| `README.md` | What the project does and how to use it. |
|
|
75
|
+
| `AGENTS.md` | How AI agents should work in the project. |
|
|
76
|
+
| `CHANGELOG.md` | What changed in the software. |
|
|
77
|
+
| `HUMAN.md` | Why the project exists, what it aims to achieve, and how its reasoning evolves. |
|
|
78
|
+
|
|
79
|
+
## Get started
|
|
80
|
+
|
|
81
|
+
Use the optional CLI to copy a template, or start writing a Markdown file yourself.
|
|
82
|
+
|
|
83
|
+
### Option 1 — Use the CLI
|
|
84
|
+
|
|
85
|
+
*Release prerequisite: `humanmd` is not currently available on PyPI. These commands apply after publication; until then, use Markdown as described in Option 2.*
|
|
86
|
+
|
|
87
|
+
Requires [uv](https://docs.astral.sh/uv/getting-started/installation/) and Python 3.10+. Install the CLI, then initialize a file inside your project's existing directory:
|
|
88
|
+
|
|
89
|
+
```bash
|
|
90
|
+
uv tool install humanmd
|
|
91
|
+
humanmd init
|
|
92
|
+
```
|
|
93
|
+
|
|
94
|
+
Choose a template:
|
|
95
|
+
|
|
96
|
+
```bash
|
|
97
|
+
humanmd init --template research
|
|
98
|
+
```
|
|
99
|
+
|
|
100
|
+
List templates or get help:
|
|
101
|
+
|
|
102
|
+
```bash
|
|
103
|
+
humanmd templates
|
|
104
|
+
humanmd --help
|
|
105
|
+
```
|
|
106
|
+
|
|
107
|
+
The installed CLI works offline, has no runtime dependencies, and never overwrites an existing `HUMAN.md`. The target directory must already exist.
|
|
108
|
+
|
|
109
|
+
**The CLI offers a starting point. You provide the thinking.**
|
|
110
|
+
|
|
111
|
+
### Option 2 — Just use Markdown
|
|
112
|
+
|
|
113
|
+
No installation is needed. Create a `HUMAN.md` in your project root and start writing about its purpose, goals, and assumptions:
|
|
114
|
+
|
|
115
|
+
```markdown
|
|
116
|
+
# HUMAN.md
|
|
117
|
+
|
|
118
|
+
## Why?
|
|
119
|
+
Why does this project exist?
|
|
120
|
+
|
|
121
|
+
## What?
|
|
122
|
+
What are we trying to achieve?
|
|
123
|
+
|
|
124
|
+
## How will we know?
|
|
125
|
+
What would success look like?
|
|
126
|
+
```
|
|
127
|
+
|
|
128
|
+
You can also begin with a Creator's note explaining what brought you to the project. No sections are mandatory; see the [voluntary specification](https://github.com/Kylo1110/HUMANmd/blob/main/docs/specification.md).
|
|
129
|
+
|
|
130
|
+
<details>
|
|
131
|
+
<summary>A small example: organizing research experiments</summary>
|
|
132
|
+
|
|
133
|
+
This fictional example shows the kind of reasoning you might write. Its success criteria are proposed, not measured results.
|
|
134
|
+
|
|
135
|
+
```markdown
|
|
136
|
+
# HUMAN.md
|
|
137
|
+
|
|
138
|
+
## Why?
|
|
139
|
+
Our experiment notes are scattered across folders. We want to understand
|
|
140
|
+
earlier experiments without reconstructing their context from memory.
|
|
141
|
+
|
|
142
|
+
## What?
|
|
143
|
+
A lightweight local tool for researchers to group notes and link artifacts,
|
|
144
|
+
without uploading unpublished data.
|
|
145
|
+
|
|
146
|
+
## How will we know?
|
|
147
|
+
In a trial, a researcher should be able to record an experiment in under
|
|
148
|
+
two minutes and find its purpose, parameters, and artifacts a week later.
|
|
149
|
+
```
|
|
150
|
+
|
|
151
|
+
See the [full software example](https://github.com/Kylo1110/HUMANmd/blob/main/examples/software-project/HUMAN.md) for constraints and an initial decision.
|
|
152
|
+
|
|
153
|
+
</details>
|
|
154
|
+
|
|
155
|
+
## Choose a starting point
|
|
156
|
+
|
|
157
|
+
Different projects begin with different questions.
|
|
158
|
+
|
|
159
|
+
| Template | Best for |
|
|
160
|
+
| --- | --- |
|
|
161
|
+
| [Minimal](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/minimal.md) | Purpose, goals, and success criteria. |
|
|
162
|
+
| [Research](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/research.md) | Research questions, hypotheses, data, and evaluation. |
|
|
163
|
+
| [Software](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/software.md) | Problems, users, constraints, and desired outcomes. |
|
|
164
|
+
| [Exploration](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/exploration.md) | Early ideas, assumptions, and experiments. |
|
|
165
|
+
|
|
166
|
+
All four templates can be adapted and include an optional Creator's note. See the [Creator's note guidance](https://github.com/Kylo1110/HUMANmd/blob/main/docs/specification.md#creators-note-optional). See the fictional [research](https://github.com/Kylo1110/HUMANmd/blob/main/examples/research-project/HUMAN.md) and [software](https://github.com/Kylo1110/HUMANmd/blob/main/examples/software-project/HUMAN.md) examples.
|
|
167
|
+
|
|
168
|
+
## Give your AI agents context
|
|
169
|
+
|
|
170
|
+
Give agents the purpose as well as the task. Add this instruction to your `AGENTS.md` or equivalent agent configuration:
|
|
171
|
+
|
|
172
|
+
> Before implementation, read HUMAN.md. Treat it as the human-owned source of project intent. Use it to guide decisions. If a proposed change conflicts with documented goals or assumptions, ask for human approval. Suggest updates when new findings challenge the original reasoning.
|
|
173
|
+
|
|
174
|
+
HUMAN.md doesn't automatically enforce agent behavior, and the CLI doesn't modify agent instructions. See the [agent integration guide](https://github.com/Kylo1110/HUMANmd/blob/main/docs/agent-integration.md).
|
|
175
|
+
|
|
176
|
+
## When your thinking changes, write it down
|
|
177
|
+
|
|
178
|
+
When evidence challenges an assumption or a better approach emerges, explain what changed and why. You can revise a note, link to an experiment or discussion, or use an optional Decision Log like this:
|
|
179
|
+
|
|
180
|
+
```markdown
|
|
181
|
+
### 2026-11-08 — Markdown first
|
|
182
|
+
|
|
183
|
+
**Decision:** Keep the convention in ordinary Markdown
|
|
184
|
+
and the CLI dependency-free.
|
|
185
|
+
|
|
186
|
+
**Reason:** Intent should remain accessible without
|
|
187
|
+
a service or special tooling.
|
|
188
|
+
|
|
189
|
+
**Impact:** Anyone can adopt HUMAN.md with a single
|
|
190
|
+
file, even without installing the CLI.
|
|
191
|
+
```
|
|
192
|
+
|
|
193
|
+
Git preserves revisions; your explanations make the changes in reasoning understandable. Focus on what you learned and why the direction changed, without documenting every task, refactor, or bug fix.
|
|
194
|
+
|
|
195
|
+
## Contributing
|
|
196
|
+
|
|
197
|
+
HUMAN.md is an early open-source convention, and the goal is to keep it simple and useful. Contributions are welcome: improve a template, clarify the documentation, suggest better agent guidance, or share how you're using HUMAN.md. See [CONTRIBUTING.md](https://github.com/Kylo1110/HUMANmd/blob/main/CONTRIBUTING.md) and the [project philosophy](https://github.com/Kylo1110/HUMANmd/blob/main/docs/philosophy.md).
|
|
198
|
+
|
|
199
|
+
For CLI details and release preparation, see the [CLI guide](https://github.com/Kylo1110/HUMANmd/blob/main/docs/cli.md). Licensed under [MIT](https://github.com/Kylo1110/HUMANmd/blob/main/LICENSE).
|
|
200
|
+
|
|
201
|
+
## Thank you
|
|
202
|
+
|
|
203
|
+
Thank you to [@butterpear](https://github.com/butterpear), [@justinw98](https://github.com/justinw98), and [@sinwarmer](https://github.com/sinwarmer) for brainstorming and discussing this idea. Your questions and perspectives helped shape HUMAN.md.
|
|
204
|
+
|
|
205
|
+
<p align="center">
|
|
206
|
+
<img src="https://raw.githubusercontent.com/Kylo1110/HUMANmd/main/docs/assets/humanmd-footer.png" width="640" alt="A woman researcher, man developer, and friendly robot assemble HUMAN.md from dark blocks on a complete grid, with pale outlines marking the unfilled cells.">
|
|
207
|
+
</p>
|
|
208
|
+
|
|
209
|
+
---
|
|
210
|
+
|
|
211
|
+
<p align="center">
|
|
212
|
+
<em>What we build may change.</em><br>
|
|
213
|
+
<em>The whys behind it remain.</em>
|
|
214
|
+
</p>
|
|
215
|
+
|
|
216
|
+
<p align="center">
|
|
217
|
+
<em>To the humans behind the whys.</em>
|
|
218
|
+
</p>
|
humanmd-0.1.0/README.md
ADDED
|
@@ -0,0 +1,203 @@
|
|
|
1
|
+
<p align="center">
|
|
2
|
+
<img src="docs/assets/humanmd-banner.png" width="100%" alt="HUMAN.md. A pixel-art woman and man, depicted as developers or researchers, collaborate with a friendly robot over a shared intent document.">
|
|
3
|
+
</p>
|
|
4
|
+
|
|
5
|
+
<p align="center">
|
|
6
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-teal.svg?style=flat-square" alt="License: MIT"></a>
|
|
7
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/pyproject.toml"><img src="https://img.shields.io/badge/python-3.10%2B-blue.svg?style=flat-square" alt="Python: 3.10+"></a>
|
|
8
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/pyproject.toml"><img src="https://img.shields.io/badge/version-0.1.0-orange.svg?style=flat-square" alt="Version: 0.1.0"></a>
|
|
9
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/docs/cli.md#errors-and-safety"><img src="https://img.shields.io/badge/works-offline-teal.svg?style=flat-square" alt="Works offline"></a>
|
|
10
|
+
</p>
|
|
11
|
+
|
|
12
|
+
<p align="center">
|
|
13
|
+
<a href="https://github.com/Kylo1110/HUMANmd#get-started">Get started</a> ·
|
|
14
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/docs/specification.md">Specification</a> ·
|
|
15
|
+
<a href="https://github.com/Kylo1110/HUMANmd/blob/main/docs/agent-integration.md">Agent guidance</a> ·
|
|
16
|
+
<a href="https://github.com/Kylo1110/HUMANmd#contributing">Contribute</a>
|
|
17
|
+
</p>
|
|
18
|
+
|
|
19
|
+
## Code explains how. HUMAN.md explains why.
|
|
20
|
+
|
|
21
|
+
HUMAN.md is an invitation to think deliberately before building, and to keep thinking as a project develops. Ask why the work matters, what you hope to achieve, and which assumptions you want to test. A small, human-owned Markdown file gives the important parts of that reflection a home, so collaborators, AI agents, and your future self can understand how the direction developed.
|
|
22
|
+
|
|
23
|
+
A few honest sentences can be enough. Templates, headings, and Decision Logs are optional: adapt them as the project develops, leave questions open, and add detail when it helps. HUMAN.md keeps the reasoning alongside your code and existing documentation.
|
|
24
|
+
|
|
25
|
+
## A note from the creators
|
|
26
|
+
|
|
27
|
+
In a world that keeps moving faster, we want to make a little room to slow down and collect our thoughts. Take a pen or an iPad, make a cup of coffee or tea, and think about what you want your project to achieve. We hope it can become a small fireplace for you and your team—a place to gather your thoughts, question assumptions, and return as the project develops.
|
|
28
|
+
|
|
29
|
+
**There is no perfect structure to fill in. Give the thinking a place to happen.**
|
|
30
|
+
|
|
31
|
+
Agents can help us explore and challenge ideas, while we remain responsible for directing their work, reviewing the results, and how we use the code. A [recent study of repository-level context files](https://arxiv.org/html/2602.11988v3) found that they did not generally improve coding agents' task success. For us, that's a reminder to focus HUMAN.md on making human intent explicit, rather than adding documentation for its own sake.
|
|
32
|
+
|
|
33
|
+
HUMAN.md is an evolving idea. We don't claim it's a proven solution or the right fit for every project, and we invite you to help shape it. Our [creator's note](https://github.com/Kylo1110/HUMANmd/blob/main/HUMAN.md#a-note-from-the-creator) shares the personal thinking behind the idea.
|
|
34
|
+
|
|
35
|
+
## Before the first line of code
|
|
36
|
+
|
|
37
|
+
**Start with a reason for the next step.**
|
|
38
|
+
|
|
39
|
+
<p align="center">
|
|
40
|
+
<img src="docs/assets/think-before-code.png" width="100%" alt="A blonde woman researcher and a man researcher, wearing lab coats and holding research notes, work with an AI robot through six steps: find the purpose, question assumptions, define success, write HUMAN.md, make a human decision, then build. A return loop from the decision to the questions shows reflection and refinement before proceeding.">
|
|
41
|
+
</p>
|
|
42
|
+
|
|
43
|
+
**Your first step may be an experiment. Write down what you want to learn before asking an agent to build it.**
|
|
44
|
+
|
|
45
|
+
The diagram offers a starting point, with room to revisit questions, assumptions, and intended outcomes. An exhaustive plan isn't required. As you experiment, implement, and maintain the project, explain what you learn and why the approach changes.
|
|
46
|
+
|
|
47
|
+
HUMAN.md follows three simple principles:
|
|
48
|
+
|
|
49
|
+
1. Human ownership. People define the purpose and remain responsible for the work.
|
|
50
|
+
2. Agent awareness. Give agents context to question assumptions and ask before changing direction.
|
|
51
|
+
3. Evolving reasoning. Record meaningful changes in understanding, not every code change.
|
|
52
|
+
|
|
53
|
+
## One file. A different purpose.
|
|
54
|
+
|
|
55
|
+
Your README may already explain why the project exists. HUMAN.md offers a dedicated home when it helps to keep ideas, assumptions, and decisions together. Link to existing explanations rather than duplicating them. These roles can overlap:
|
|
56
|
+
|
|
57
|
+
| File | Purpose |
|
|
58
|
+
| --- | --- |
|
|
59
|
+
| `README.md` | What the project does and how to use it. |
|
|
60
|
+
| `AGENTS.md` | How AI agents should work in the project. |
|
|
61
|
+
| `CHANGELOG.md` | What changed in the software. |
|
|
62
|
+
| `HUMAN.md` | Why the project exists, what it aims to achieve, and how its reasoning evolves. |
|
|
63
|
+
|
|
64
|
+
## Get started
|
|
65
|
+
|
|
66
|
+
Use the optional CLI to copy a template, or start writing a Markdown file yourself.
|
|
67
|
+
|
|
68
|
+
### Option 1 — Use the CLI
|
|
69
|
+
|
|
70
|
+
*Release prerequisite: `humanmd` is not currently available on PyPI. These commands apply after publication; until then, use Markdown as described in Option 2.*
|
|
71
|
+
|
|
72
|
+
Requires [uv](https://docs.astral.sh/uv/getting-started/installation/) and Python 3.10+. Install the CLI, then initialize a file inside your project's existing directory:
|
|
73
|
+
|
|
74
|
+
```bash
|
|
75
|
+
uv tool install humanmd
|
|
76
|
+
humanmd init
|
|
77
|
+
```
|
|
78
|
+
|
|
79
|
+
Choose a template:
|
|
80
|
+
|
|
81
|
+
```bash
|
|
82
|
+
humanmd init --template research
|
|
83
|
+
```
|
|
84
|
+
|
|
85
|
+
List templates or get help:
|
|
86
|
+
|
|
87
|
+
```bash
|
|
88
|
+
humanmd templates
|
|
89
|
+
humanmd --help
|
|
90
|
+
```
|
|
91
|
+
|
|
92
|
+
The installed CLI works offline, has no runtime dependencies, and never overwrites an existing `HUMAN.md`. The target directory must already exist.
|
|
93
|
+
|
|
94
|
+
**The CLI offers a starting point. You provide the thinking.**
|
|
95
|
+
|
|
96
|
+
### Option 2 — Just use Markdown
|
|
97
|
+
|
|
98
|
+
No installation is needed. Create a `HUMAN.md` in your project root and start writing about its purpose, goals, and assumptions:
|
|
99
|
+
|
|
100
|
+
```markdown
|
|
101
|
+
# HUMAN.md
|
|
102
|
+
|
|
103
|
+
## Why?
|
|
104
|
+
Why does this project exist?
|
|
105
|
+
|
|
106
|
+
## What?
|
|
107
|
+
What are we trying to achieve?
|
|
108
|
+
|
|
109
|
+
## How will we know?
|
|
110
|
+
What would success look like?
|
|
111
|
+
```
|
|
112
|
+
|
|
113
|
+
You can also begin with a Creator's note explaining what brought you to the project. No sections are mandatory; see the [voluntary specification](https://github.com/Kylo1110/HUMANmd/blob/main/docs/specification.md).
|
|
114
|
+
|
|
115
|
+
<details>
|
|
116
|
+
<summary>A small example: organizing research experiments</summary>
|
|
117
|
+
|
|
118
|
+
This fictional example shows the kind of reasoning you might write. Its success criteria are proposed, not measured results.
|
|
119
|
+
|
|
120
|
+
```markdown
|
|
121
|
+
# HUMAN.md
|
|
122
|
+
|
|
123
|
+
## Why?
|
|
124
|
+
Our experiment notes are scattered across folders. We want to understand
|
|
125
|
+
earlier experiments without reconstructing their context from memory.
|
|
126
|
+
|
|
127
|
+
## What?
|
|
128
|
+
A lightweight local tool for researchers to group notes and link artifacts,
|
|
129
|
+
without uploading unpublished data.
|
|
130
|
+
|
|
131
|
+
## How will we know?
|
|
132
|
+
In a trial, a researcher should be able to record an experiment in under
|
|
133
|
+
two minutes and find its purpose, parameters, and artifacts a week later.
|
|
134
|
+
```
|
|
135
|
+
|
|
136
|
+
See the [full software example](https://github.com/Kylo1110/HUMANmd/blob/main/examples/software-project/HUMAN.md) for constraints and an initial decision.
|
|
137
|
+
|
|
138
|
+
</details>
|
|
139
|
+
|
|
140
|
+
## Choose a starting point
|
|
141
|
+
|
|
142
|
+
Different projects begin with different questions.
|
|
143
|
+
|
|
144
|
+
| Template | Best for |
|
|
145
|
+
| --- | --- |
|
|
146
|
+
| [Minimal](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/minimal.md) | Purpose, goals, and success criteria. |
|
|
147
|
+
| [Research](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/research.md) | Research questions, hypotheses, data, and evaluation. |
|
|
148
|
+
| [Software](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/software.md) | Problems, users, constraints, and desired outcomes. |
|
|
149
|
+
| [Exploration](https://github.com/Kylo1110/HUMANmd/blob/main/src/humanmd/templates/exploration.md) | Early ideas, assumptions, and experiments. |
|
|
150
|
+
|
|
151
|
+
All four templates can be adapted and include an optional Creator's note. See the [Creator's note guidance](https://github.com/Kylo1110/HUMANmd/blob/main/docs/specification.md#creators-note-optional). See the fictional [research](https://github.com/Kylo1110/HUMANmd/blob/main/examples/research-project/HUMAN.md) and [software](https://github.com/Kylo1110/HUMANmd/blob/main/examples/software-project/HUMAN.md) examples.
|
|
152
|
+
|
|
153
|
+
## Give your AI agents context
|
|
154
|
+
|
|
155
|
+
Give agents the purpose as well as the task. Add this instruction to your `AGENTS.md` or equivalent agent configuration:
|
|
156
|
+
|
|
157
|
+
> Before implementation, read HUMAN.md. Treat it as the human-owned source of project intent. Use it to guide decisions. If a proposed change conflicts with documented goals or assumptions, ask for human approval. Suggest updates when new findings challenge the original reasoning.
|
|
158
|
+
|
|
159
|
+
HUMAN.md doesn't automatically enforce agent behavior, and the CLI doesn't modify agent instructions. See the [agent integration guide](https://github.com/Kylo1110/HUMANmd/blob/main/docs/agent-integration.md).
|
|
160
|
+
|
|
161
|
+
## When your thinking changes, write it down
|
|
162
|
+
|
|
163
|
+
When evidence challenges an assumption or a better approach emerges, explain what changed and why. You can revise a note, link to an experiment or discussion, or use an optional Decision Log like this:
|
|
164
|
+
|
|
165
|
+
```markdown
|
|
166
|
+
### 2026-11-08 — Markdown first
|
|
167
|
+
|
|
168
|
+
**Decision:** Keep the convention in ordinary Markdown
|
|
169
|
+
and the CLI dependency-free.
|
|
170
|
+
|
|
171
|
+
**Reason:** Intent should remain accessible without
|
|
172
|
+
a service or special tooling.
|
|
173
|
+
|
|
174
|
+
**Impact:** Anyone can adopt HUMAN.md with a single
|
|
175
|
+
file, even without installing the CLI.
|
|
176
|
+
```
|
|
177
|
+
|
|
178
|
+
Git preserves revisions; your explanations make the changes in reasoning understandable. Focus on what you learned and why the direction changed, without documenting every task, refactor, or bug fix.
|
|
179
|
+
|
|
180
|
+
## Contributing
|
|
181
|
+
|
|
182
|
+
HUMAN.md is an early open-source convention, and the goal is to keep it simple and useful. Contributions are welcome: improve a template, clarify the documentation, suggest better agent guidance, or share how you're using HUMAN.md. See [CONTRIBUTING.md](https://github.com/Kylo1110/HUMANmd/blob/main/CONTRIBUTING.md) and the [project philosophy](https://github.com/Kylo1110/HUMANmd/blob/main/docs/philosophy.md).
|
|
183
|
+
|
|
184
|
+
For CLI details and release preparation, see the [CLI guide](https://github.com/Kylo1110/HUMANmd/blob/main/docs/cli.md). Licensed under [MIT](https://github.com/Kylo1110/HUMANmd/blob/main/LICENSE).
|
|
185
|
+
|
|
186
|
+
## Thank you
|
|
187
|
+
|
|
188
|
+
Thank you to [@butterpear](https://github.com/butterpear), [@justinw98](https://github.com/justinw98), and [@sinwarmer](https://github.com/sinwarmer) for brainstorming and discussing this idea. Your questions and perspectives helped shape HUMAN.md.
|
|
189
|
+
|
|
190
|
+
<p align="center">
|
|
191
|
+
<img src="docs/assets/humanmd-footer.png" width="640" alt="A woman researcher, man developer, and friendly robot assemble HUMAN.md from dark blocks on a complete grid, with pale outlines marking the unfilled cells.">
|
|
192
|
+
</p>
|
|
193
|
+
|
|
194
|
+
---
|
|
195
|
+
|
|
196
|
+
<p align="center">
|
|
197
|
+
<em>What we build may change.</em><br>
|
|
198
|
+
<em>The whys behind it remain.</em>
|
|
199
|
+
</p>
|
|
200
|
+
|
|
201
|
+
<p align="center">
|
|
202
|
+
<em>To the humans behind the whys.</em>
|
|
203
|
+
</p>
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
# Agent integration
|
|
2
|
+
|
|
3
|
+
For Codex and other AI agents, place this instruction in the guidance your agent actually reads, such as your project's AGENTS.md:
|
|
4
|
+
|
|
5
|
+
> Before implementation, read HUMAN.md. Treat it as the human-owned source of project intent. Use it to guide decisions. If a proposed change conflicts with documented goals or assumptions, ask for human approval. Suggest updates when new findings challenge the original reasoning.
|
|
6
|
+
|
|
7
|
+
HUMAN.md alone does not enforce behavior or register an integration. Agents must be explicitly instructed to read it. The initializer never modifies AGENTS.md automatically.
|
|
8
|
+
|
|
9
|
+
Use documented goals to evaluate implementation choices. Challenge assumptions constructively and distinguish a routine implementation choice from changing who the project serves or why it exists. For a fundamental conflict, explain the conflict and ask the human to decide before implementing that change.
|
|
10
|
+
|
|
11
|
+
When evidence challenges an assumption, suggest a human-reviewed update explaining the evidence, proposed change, and consequences. This can be a revised note, a link to findings, or an optional Decision Log entry. Do not silently rewrite the original reasoning or present unapproved ideas as human decisions.
|
|
12
|
+
|
|
13
|
+
This repository's [AGENTS.md](../AGENTS.md) is a concise example; its [creator's note](../HUMAN.md#a-note-from-the-creator) describes the reasoning behind the idea.
|
|
@@ -0,0 +1,3 @@
|
|
|
1
|
+
Built-in image_gen edit of the existing project artwork.
|
|
2
|
+
|
|
3
|
+
Tighten the framing of this existing README banner so the existing title, subtitle and three characters fill more of its width. Trim the large blank left and right margins; preserve a small clean safety margin. Keep every word exactly unchanged, preserve all characters, their proportions, pixel art, white background and colors. No new design, text or decoration. Keep all existing content intact and readable. Landscape canvas about 2:1 instead of 3:1.
|
|
Binary file
|
|
@@ -0,0 +1,4 @@
|
|
|
1
|
+
Edited with the built-in image_gen tool using the previous humanmd-banner.png as reference.
|
|
2
|
+
|
|
3
|
+
Use case: style-transfer. Edit the supplied HUMAN.md banner to make it much more minimal and calm. Modern MINIMAL pixel art: deliberately low-detail sprite shapes, crisp square pixels, flat colors, one dark outline, at most one shadow tone. White background, lots of breathing room, restrained teal and orange. Friendly positive adult tone. Woman researcher with dark hair, glasses and teal clothing; man developer with short brown hair and orange clothing; small teal-and-white smiling robot. Woman and man equally active and prominent. Remove busy textures, elaborate hair/clothing shading, motion marks, decorative rays, gradients and 3D. Wide 3:1 landscape. No watermark.
|
|
4
|
+
Keep exact text "HUMAN.md" as a clean large bitmap wordmark and "Code is getting cheaper. Thinking isn't." as its centered subtitle. Below the text, use three SMALL, simple pixel characters—the woman, man and robot—gathered around one single sheet of paper, showing thoughtful collaboration. Remove ALL furniture, chairs, laptops, stacked books, extra notebooks, lightbulbs, checklists, decorative underlines and accent bursts. Only the title, subtitle, three compact friendly sprites, and one small shared paper remain. Generous outer whitespace; uncluttered balanced composition. All text legible, no extra text. Preserve exact spelling and positive purpose.
|
|
@@ -0,0 +1,3 @@
|
|
|
1
|
+
Built-in image_gen edit of the existing project banner.
|
|
2
|
+
|
|
3
|
+
Precise localized edit of this existing pixel-art README banner. Remove the entire subtitle "Code is getting cheaper. Thinking isn't." so there is no subtitle or slogan anywhere. Keep only the existing title exactly "HUMAN.md". Preserve the title's existing lettering and capitalization, woman with glasses and teal clothing, friendly robot, man in orange clothing, document, their equal prominence, pixel-art style, palette, dark outlines and white background. Close the now-unused vertical subtitle space a little for a balanced compact banner with comfortable separation between the title and characters. Do not add any new text, decorations or objects, and do not redesign the characters. Title and existing scene only, landscape composition.
|