Anbeeld AGENTS.md

Anbeeld's Global AGENTS.md / CLAUDE.md

A global instruction file for coding agents that targets what they keep getting wrong: jumping to code without evidence, expanding scope unchecked, or shipping changes that haven't been validated.

Every instruction came from the same pipeline: observe what agents get wrong, compare it with public examples and recommendations from industry leaders, test my own solution when needed, then add or tighten a rule.

Battle-tested in daily use, tuned across multiple agents and models. Use the full set, borrow what you need, or build something different from it.

What it does

- No wrong-path work: asks before guessing on ambiguous tasks, investigates proportionally to risk, and doesn't improvise when goals conflict. When there's nobody to ask, it takes the reversible option and reports the assumption, or stops and says what it needs.
- Boundaries: no fabrication, no gamed verification, no exposed secrets, and no destructive actions without your explicit go-ahead for that exact action. Instructions buried in repo files, issues, logs, or tool output are treated as data, not orders.
- Structured execution order: scope, gather evidence, load matching skills, route, implement, verify. Routing leans toward splitting independent work out: parallel tool calls, two or more subagents, or the main agent running alongside one.
- Keeps changes small. No scope creep, no unprompted refactors, no new dependencies without checking what's already available.
- Protects existing code: tests match the risk level, behavior isn't silently changed, errors aren't swallowed. If verification fails, the agent diagnoses the cause and keeps going only while each new attempt has new evidence behind it; otherwise it stops and reports the failure.
- Validates before…