this repo has no description
agent skills, here we go
+50
+49
ai/skills/dev-philosophy.md
+49
ai/skills/dev-philosophy.md
···
1
+
# Software Development Philosophy
2
+
3
+
This skill defines how I approach software development. Apply these principles whenever working on code.
4
+
5
+
## Core Principle: Ruthless Simplicity
6
+
7
+
- Write the minimum code necessary to solve the problem
8
+
- Every line must justify its existence
9
+
- Minimize maintenance burden — less code means less to understand, review, debug
10
+
- Do not over-engineer unless there is a clear, proven need
11
+
- Avoid abstractions until they're obviously necessary
12
+
13
+
## Approach to New Work
14
+
15
+
1. **Clarify the problem first** — understand exactly what needs to be solved before writing any code
16
+
2. **Design before implementing** — think through data structures, functions, and interfaces at a high level
17
+
3. **Iterate on the design** — refine until the structure is solid and any unnecessary complexity is removed
18
+
4. **Then implement** — only after the above steps are complete
19
+
20
+
## Testing
21
+
22
+
- Always prefer automated tests that verify behavior deterministically
23
+
- Write code that is testable
24
+
- Prefer tests that run against real conditions (real DB like sqlite, end-to-end) over heavy mocking
25
+
- Keep tests themselves simple — they are code too and carry maintenance burden
26
+
- Tests are part of "done"
27
+
28
+
## Technology Preferences
29
+
30
+
- Prefer lightweight tools (sqlite over heavy databases)
31
+
- Prefer simple HTML over heavy UI frameworks
32
+
- Minimize client-side complexity
33
+
- CLIs and JSON APIs are preferred interfaces
34
+
- Avoid unnecessary dependencies
35
+
36
+
## Boundaries
37
+
38
+
- **Stay focused**: Only implement what is directly requested — no "helpful" extras or unrequested improvements
39
+
- **No commits without permission**: Do not commit unless explicitly asked. A reminder that changes are ready to commit is acceptable.
40
+
- **No pushing**: Never push to remote without explicit instruction
41
+
- **Lint and format**: Ensure code passes linting and formatting before considering work complete
42
+
43
+
## What "Done" Looks Like
44
+
45
+
- The problem is solved with minimal code
46
+
- Automated tests verify the behavior
47
+
- Code is linted and formatted
48
+
- Diff has been reviewed (by the user, not the agent)
49
+
- Ready to commit, but waiting for explicit go-ahead