claude-code-for-hydrology
CLAUDE.md67 starsChanged 6 months agoMITView on GitHub
Structured exercises for learning Claude Code best practices, using hydrology-themed Python code. Each numbered folder is a self-contained exercise. Read the folder's README.md before starting. Use python3 on systems where python defaults to Python 2.
In this repository3 other files it gives its agents.
- flow-reportSkill
- hydro-contextSkill
- run-testsSkill
The file
We split this CLAUDE.md into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Use python3 on systems where python defaults to Python 2.
- Distinctive: line 30
Read the README.md in the exercise folder
- Distinctive: line 37
All exercises use hydrology/water science Python code (streamflow, drought indices, model metrics)
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Claude Code Best Practices — Learning Project
Structured exercises for learning Claude Code best practices, using hydrology-themed Python code.
Each numbered folder is a self-contained exercise. Read the folder's README.md before starting.
Python Environment
python -m pytest <exercise_folder>/ -vUse python3 on systems where python defaults to Python 2.
Exercise Index
| Folder | Best Practice |
01_explore_plan_code/ | Explore → Plan → Code with plan mode |
02_specific_context/ | Reference specific files and symptoms in prompts |
03_verify_your_work/ | Give Claude tests so it can verify its own output |
04_init_claude_md/ | Use /init to create CLAUDE.md for persistent project context |
05_skills/ | Create and use skills for domain knowledge (/hydro-context) and repeatable workflows (/flow-report) |
06_subagent_review/ | Orchestrate coder and reviewer subagents in sequence; parent controls context isolation and loops until review passes |
07_aws_cli_workflow/ | Use AWS CLI with Claude to download public hydrological datasets |
08_mcp_usgs_gauge/ | Use MCP fetch server to query the USGS NWIS API and implement correct parsing code |
09_parallel_fanout/ | Spawn subagents in parallel for multi-configuration analysis; aggregate with a reporter subagent |
Workflow for Each Exercise
- Read the
README.mdin the exercise folder - Try the "before" prompt — observe what Claude produces
- Clear context with
/clear - Try the "after" prompt — compare the results
Notes
- All exercises use hydrology/water science Python code (streamflow, drought indices, model metrics)
- The
.claude/folder contains a pre-built skill (/run-tests) and a subagent (code-reviewer) to examine as examples
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.