Claude Code Agent Development
anthropics/claude-plugins-official
Explains how to write agents for Claude Code plugins: the markdown file with YAML frontmatter, trigger descriptions, model and color settings, and system prompt design.
Meta-skill that generates Claude Code agent definition files (.md).
Install Subagent Creator
$ npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install idoforgod/Dissertation-Simulator-AgenticWorkflow subagent-creator --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/subagent-creator .claude/skills/subagent-creator && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "subagent-creator" agent skill from https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creator into .claude/skills/subagent-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "subagent-creator", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creatorType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install idoforgod/Dissertation-Simulator-AgenticWorkflow subagent-creator --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/subagent-creator .agents/skills/subagent-creator && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "subagent-creator" agent skill from https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creator into .agents/skills/subagent-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "subagent-creator", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install idoforgod/Dissertation-Simulator-AgenticWorkflow subagent-creator --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/subagent-creator .cursor/skills/subagent-creator && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "subagent-creator" agent skill from https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creator into .cursor/skills/subagent-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "subagent-creator", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git --path .claude/skills/subagent-creator--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install idoforgod/Dissertation-Simulator-AgenticWorkflow subagent-creator --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/subagent-creator .gemini/skills/subagent-creator && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "subagent-creator" agent skill from https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creator into .gemini/skills/subagent-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "subagent-creator", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install idoforgod/Dissertation-Simulator-AgenticWorkflow subagent-creatorInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/subagent-creator .github/skills/subagent-creator && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "subagent-creator" agent skill from https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creator into .github/skills/subagent-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "subagent-creator", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install idoforgod/Dissertation-Simulator-AgenticWorkflow subagent-creator --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/subagent-creator .opencode/skills/subagent-creator && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "subagent-creator" agent skill from https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflow/tree/main/.claude/skills/subagent-creator into .opencode/skills/subagent-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "subagent-creator", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
subagent-creator- GitHub stars
- 108
- Token cost
- ~1.9k tokens
- SKILL.md length
- 737 words
- Files
- 5 (incl. references)
- Skills in repo
- 4
- Repo updated
- First seen
- Licence
- MIT
At a glance
Meta-skill that generates Claude Code agent definition files (.md).
- Works in 8 steps: Agent Design Analysis → Model Selection → Tool Selection → …
- Asked to create an agent
- SKILL.md covers When to Use, Inherited DNA, Agent File Schema and Generation Protocol, plus 3 more sections
- Calls python3
What it does
Subagent Creator is an agent skill from idoforgod/Dissertation-Simulator-AgenticWorkflow. Meta-skill that generates Claude Code agent definition files (.md). Creates specialized agents with proper tool access, model selection, and behavioral instructions. Use when asked to "create an agent", "make a sub-agent", or "generate agent for X".
Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `references/agent-template-guide.md`, `references/gra-schema-reference.md` and `references/output-path-conventions.md`).
It sits in Agent Workflows, covering Subagents. The repository describes itself as: AI 에이전트 58개가 협업하는 211-step 박사 논문 연구 시뮬레이션 시스템. AgenticWorkflow 프레임워크에서 태어난 자식 시스템. The licence is MIT.
When your agent uses it
- Asked to create an agent
- Make a sub-agent
- Generate agent for X
Example prompts
- “create an agent”
- “make a sub-agent”
- “generate agent for X”
- “/subagent-creator”
Requirements
- Python 3
Workflow steps
8 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit f2205f5. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
python3
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Subagent Creator loads about 1.9k tokens when it runs, and up to ~6.1k if it reads all its reference files. Until then it costs about 67 tokens; SKILL.md has 737 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from idoforgod/Dissertation-Simulator-AgenticWorkflow at commit f2205f5, republished under its MIT licence (© idoforgod). 737 words, ~1,900 tokens.
Download SKILL.mdSave it as
.claude/skills/subagent-creator/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.- name
- subagent-creator
- description
- Meta-skill that generates Claude Code agent definition files (.md). Creates specialized agents with proper tool access, model selection, and behavioral instructions. Use when asked to "create an agent", "make a sub-agent", or "generate agent for X".
Sub-Agent Creator — Meta-Skill for Generating Agent Definitions
A meta-skill that produces Claude Code agent .md files. It designs agents
with appropriate expertise, tool access, behavioral rules, and quality
requirements following AgenticWorkflow conventions.
When to Use
- User asks to create a new agent for a specific task
- System needs to generate agents for a workflow (e.g., thesis agents)
- Batch creation of multiple related agents
Inherited DNA
This meta-skill inherits the AgenticWorkflow genome. As a skill that generates agents, it must itself embody the DNA it enforces.
| DNA Component | Expression in subagent-creator |
|---|---|
| Absolute Criteria 1 (Quality) | Generated agents use optimal model selection (opus for research, sonnet for utility) |
| Absolute Criteria 2 (SOT) | Generated agents respect single-writer SOT pattern; guard_sot_write.py compatibility |
| Absolute Criteria 3 (CCP) | Research agents include GRA compliance; utility agents document CCP exemption rationale |
| English-First | All agent instructions are in English |
| P1 Compliance | Research agents include GroundedClaim schema + Hallucination Firewall |
| Quality Gates | Research agents integrate with validate_grounded_claim.py PostToolUse hook |
Agent File Schema
Agent definitions live in .claude/agents/ and follow this structure:
---
name: {agent-name} # kebab-case
description: {brief description}
model: opus # opus | sonnet | haiku
tools: Read, Write, Glob, Grep # comma-separated tool list
maxTurns: {N} # max reasoning turns (default: 20)
memory: project # project | none
---
{Agent behavioral instructions in English}Generation Protocol
Step 1: Agent Design Analysis
Determine the agent's requirements:
- Primary task: What is the agent's core responsibility?
- Expertise domain: What specialized knowledge does it need?
- Input/Output: What does it receive and produce?
- Quality criteria: What makes a good output for this agent?
- Tool needs: Which tools does it require?
Step 2: Model Selection
Select the appropriate model based on task complexity:
| Model | Use When | Examples |
|---|---|---|
opus | Complex analysis, synthesis, critical reasoning | Thesis writer, critical reviewer, synthesis agent |
sonnet | Structured tasks, search, data processing | Literature searcher, formatting specialist |
haiku | Simple, repetitive tasks | File validation, format checking |
Default to opus when quality is the absolute criterion (Absolute Criteria 1).
Step 3: Tool Selection
Assign tools based on the agent's needs:
| Tool | When to Include |
|---|---|
Read | Agent needs to read files (almost always) |
Write | Agent produces output files |
Glob | Agent needs to find files by pattern |
Grep | Agent needs to search file contents |
Bash | Agent needs to run commands (use sparingly) |
WebSearch | Agent needs to search the web |
WebFetch | Agent needs to fetch web content |
Agent | Agent needs to delegate to sub-agents |
Step 4: Generate Agent Definition
Write the agent .md file with:
- Frontmatter: name, description, model, tools, maxTurns, memory
- Role definition: "You are a [role] specializing in [domain]."
- Task instructions: Step-by-step protocol for the agent's work
- Output format: Exact specification of expected output structure
- Quality rules: Domain-specific quality requirements
- GRA compliance (if research agent): GroundedClaim schema, Hallucination Firewall rules
Show full SKILL.md (317 more words)Show less
Step 5: Context Isolation Assessment
Determine if commands invoking this agent should use context: fork:
| Factor | Inline (no fork) | Fork recommended |
|---|---|---|
| Agent is part of orchestration flow | ✅ | ❌ |
| Agent writes to SOT | ✅ (orchestrator only) | ❌ never |
| Agent does independent analysis/production | ❌ | ✅ |
| Agent needs Bash for P1 validation scripts | Fork requires Bash in tool list | Check tool compatibility |
| Agent's work would pollute main context | ❌ | ✅ |
If fork is recommended, note this in the agent's documentation:
## Fork Compatibility
This agent is safe for `context: fork` invocation. It:
- Reads SOT but never writes to it
- Produces independent output files at {output_path}
- Does not require Bash / Does require Bash (specify)Most thesis workflow agents should NOT be forked — they are invoked by thesis-orchestrator within Agent Teams, which already provides context isolation.
Step 6: GRA Integration (Research Agents Only)
For agents that produce research claims, add:
## GRA Compliance
All claims must follow the GroundedClaim schema:
- **id**: "{CLAIM_PREFIX}-{NNN}" (e.g., "LS-001")
- **claim_type**: FACTUAL | EMPIRICAL | THEORETICAL | METHODOLOGICAL | INTERPRETIVE | SPECULATIVE
- **sources**: At least one PRIMARY or SECONDARY source with reference and DOI
- **confidence**: 0-100 score
- **effect_size**: When applicable (statistical findings)
- **uncertainty**: Explicit limitation statement
### Hallucination Firewall
- BLOCK: "all studies agree", "100%", "no exceptions"
- REQUIRE_SOURCE: Any statistical claim (p-values, effect sizes)
- SOFTEN: "certainly", "obviously", "clearly" → add hedging
- VERIFY: "it is known that" → add citationStep 7: Validate Agent Definition
Verify the generated agent:
- Frontmatter has all required fields
- name matches filename (kebab-case)
- Instructions are in English (AI performance)
- Output format is clearly specified
- Tool list matches actual needs
- GRA compliance section present (if research agent)
- No placeholder content
- Fork Safety Cross-Validation (if fork-compatible in Step 5):
Any command/skill using
agent: {this-agent}withcontext: forkmust pass:python3 .claude/hooks/scripts/validate_fork_safety.py --file <command.md> --project-dir <project-root>Validates FS-3 (Bash dependency vs agent tools) and FS-5 (agent existence).
Step 8: Register Agent
Output:
- Agent file location:
.claude/agents/{name}.md - How to invoke:
@{name}in prompts or via Agent tool - Claim prefix (if GRA agent):
{PREFIX}
Language Rules
- Agent instructions (body): English — AI performance optimization
- Description (frontmatter): English — for agent matching
- Output instructions for user-facing text: Include Korean translation directive
Batch Creation
When creating multiple related agents:
- Design all agents together for consistency
- Ensure claim prefixes are unique across the set
- Define inter-agent dependencies explicitly
- Verify no overlapping responsibilities
Quality Checklist
- Agent follows AgenticWorkflow conventions
- Model selection justified by task complexity
- Tool list is minimal but sufficient
- Instructions are specific and actionable
- GRA compliance complete (for research agents)
- Claim prefix unique (for GRA agents)
- Fork compatibility assessed (Step 5) and documented if applicable
- No hardcoded file paths
© idoforgod, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 4 other files (references) in .claude/skills/subagent-creator of idoforgod/Dissertation-Simulator-AgenticWorkflow.
- SKILL.md
- references/agent-template-guide.md
- references/gra-schema-reference.md
- references/output-path-conventions.md
- references/tier-routing-guide.md
Open the folder on GitHubat commit f2205f5
Compare with similar skills
Subagent Creator next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Subagent Creator this skillidoforgod/Dissertation-Simulator-AgenticWorkflow | 108 | — | ~1.9k | Automated safety check: Pass | MIT | |
| Claude Code Agent Developmentanthropics/claude-plugins-official | 38k | 8 repos | ~2.8k | Automated safety check: Pass | Apache-2.0 | |
| Subagent Driven DevelopmentAsvarox/allkaraoke | 261 | 38 repos | ~1.2k | Automated safety check: Pass | None | |
| Dispatching Parallel Agentsultralisp/ultralisp | 258 | 41 repos | ~1.5k | Automated safety check: Pass | None | |
| Paseo Advisor Second Opiniongetpaseo/paseo | 20k | 1 repos | ~756 | Automated safety check: Pass | Custom licence | |
| Task Observerrebelytics/one-skill-to-rule-them-all | 3.2k | 1 repos | ~12k | Automated safety check: Pass | CC-BY-4.0 |
Similar skills
- Official
Subagent Driven Development
Asvarox/allkaraoke
A skill your agent uses when executing implementation plans with independent tasks in the current session
Dispatching Parallel Agents
ultralisp/ultralisp
A skill your agent uses when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Paseo Advisor Second Opinion
getpaseo/paseo
Launches one separate agent through Paseo to give a second opinion on the current task, with a self-contained briefing and no permission to edit files.
Task Observer
rebelytics/one-skill-to-rule-them-all
Monitors task execution for skill improvement opportunities.
O2 Review Loop
openobserve/openobserve
Splits a change into planner, coder and independent reviewer roles: you confirm a spec, a subagent implements it, and a separate reviewer checks each round's local WIP commit.
More from idoforgod/Dissertation-Simulator-AgenticWorkflow
Skill Creator
idoforgod/Dissertation-Simulator-AgenticWorkflow
Meta-skill that generates new Claude Code skills. An agent skill from idoforgod/Dissertation-Simulator-AgenticWorkflow.
Doctoral Writing
idoforgod/Dissertation-Simulator-AgenticWorkflow
박사급 학위 논문 글쓰기 스킬. An agent skill from idoforgod/Dissertation-Simulator-AgenticWorkflow.
Workflow Generator
idoforgod/Dissertation-Simulator-AgenticWorkflow
Automated workflow (workflow.md) generator for Claude Code. An agent skill from idoforgod/Dissertation-Simulator-AgenticWorkflow.
Related
Categories
Questions about Subagent Creator
What does Subagent Creator do?
Meta-skill that generates Claude Code agent definition files (.md). Subagent Creator is an agent skill from idoforgod/Dissertation-Simulator-AgenticWorkflow.md).
When should I use Subagent Creator?
Subagent Creator fits situations like: asked to create an agent; make a sub-agent; generate agent for X.
How do I install Subagent Creator in Claude Code?
Run `npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a claude-code`. Or copy the skill folder (.claude/skills/subagent-creator in idoforgod/Dissertation-Simulator-AgenticWorkflow) into .claude/skills/subagent-creator in your project. Claude Code loads it when a task matches its description.
How do I install Subagent Creator in Codex?
Run `npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a codex`. Or copy the skill folder (.claude/skills/subagent-creator in idoforgod/Dissertation-Simulator-AgenticWorkflow) into .agents/skills/subagent-creator in your project. Codex loads it when a task matches its description.
Can I use Subagent Creator in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add idoforgod/Dissertation-Simulator-AgenticWorkflow --skill subagent-creator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/subagent-creator, .gemini/skills/subagent-creator, .github/skills/subagent-creator and .opencode/skills/subagent-creator in your project.
What does Subagent Creator need to run?
Going by SKILL.md and its folder, Subagent Creator needs the command-line tools its instructions call (python3). Our summary lists: Python 3.
Does Subagent Creator access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Subagent Creator safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Subagent Creator use?
Subagent Creator is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Subagent Creator use?
About 1.9k tokens (SKILL.md is roughly 7.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 4.2k tokens, read only when the agent opens those files.
What are the alternatives to Subagent Creator?
Skills that share tags, products or a category with Subagent Creator: Claude Code Agent Development (anthropics/claude-plugins-official, 38k stars), Subagent Driven Development (Asvarox/allkaraoke, 261 stars), Dispatching Parallel Agents (ultralisp/ultralisp, 258 stars) and Paseo Advisor Second Opinion (getpaseo/paseo, 20k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Subagent Creator?
idoforgod (a GitHub user) maintains it in idoforgod/Dissertation-Simulator-AgenticWorkflow, which has 108 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on June 4, 2026.
Source: idoforgod/Dissertation-Simulator-AgenticWorkflow on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.