Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the browser with native browser sessions (multi-harness + multi-model) like OpenClaw (voice supported + ToS friendly)
Python
Align AGENTS.md and CLAUDE.md with architecture plan principles
Cursor Agentdb1ae0fea957
diff --git a/AGENTS.md b/AGENTS.mdindex 82a5c33c..33c1e2fd 100644--- a/AGENTS.md+++ b/AGENTS.md@@ -18,16 +18,27 @@- Goal: Zero-defect, root-cause-oriented engineering for bugs and test-driven engineering for new features. Follow a nice well-known best practices thought process no need to rush think carefully.- Code: You must aim to write the simplest code possible keeping the code base minimal and modular according to best practices to prevent complicating things+## ARCHITECTURE PRINCIPLES (see PLAN.md)+- **Shared utilities**: Extract common logic into shared packages (e.g. `providers/common/`). Do not have one provider import from another provider's utils.+- **DRY**: Extract shared base classes to eliminate duplication. Prefer composition over copy-paste.+- **Encapsulation**: Use accessor methods for internal state (e.g. `set_current_task()`), not direct `_attribute` assignment from outside.+- **Provider-specific config**: Keep provider-specific fields (e.g. `nim_settings`) in provider constructors, not in the base `ProviderConfig`.+- **Dead code**: Remove unused code, legacy systems, and hardcoded values. Use settings/config instead of literals (e.g. `settings.provider_type` not `"nvidia_nim"`).+- **Performance**: Use list accumulation for strings (not `+=` in loops), cache env vars at init, prefer iterative over recursive when stack depth matters.+- **Platform-agnostic naming**: Use generic names (e.g. `PLATFORM_EDIT`) not platform-specific ones (e.g. `TELEGRAM_EDIT`) in shared code.+- **No type ignores**: Do not add `# type: ignore` or `# ty: ignore`. Fix the underlying type issue.+- **Backward compatibility**: When moving modules, add re-exports from old locations so existing imports keep working.+## COGNITIVE WORKFLOW1. ANALYZE: Read relevant files if you have not already. Do not guess.-2. PLAN: Use thinking mode to map out the logic. Identify the root cause or required changes.+2. PLAN: Use thinking mode to map out the logic. Identify the root cause or required changes. Order changes by dependency.3. EXECUTE: Fix the cause, not the symptom. Execute smartly and carefully4. VERIFY: Run tests or linting. Confirm the fix via logs or output.5. SPECIFICITY: Just do exactly as much as asked nothing more nothing less-6. PROPAGATION: Making changes has impacts across files so proapagate changes correctly+6. PROPAGATION: Making changes has impacts across files so propagate changes correctly## SUMMARY STANDARDS-- Summaries must be technical and granular.+- Summaries must be technical and granular.- Include: [Files Changed], [Logic Altered], [Verification Method], [Residual Risks].## TOOLSdiff --git a/CLAUDE.md b/CLAUDE.mdindex 82a5c33c..33c1e2fd 100644--- a/CLAUDE.md+++ b/CLAUDE.md@@ -18,16 +18,27 @@- Goal: Zero-defect, root-cause-oriented engineering for bugs and test-driven engineering for new features. Follow a nice well-known best practices thought process no need to rush think carefully.- Code: You must aim to write the simplest code possible keeping the code base minimal and modular according to best practices to prevent complicating things+## ARCHITECTURE PRINCIPLES (see PLAN.md)+- **Shared utilities**: Extract common logic into shared packages (e.g. `providers/common/`). Do not have one provider import from another provider's utils.+- **DRY**: Extract shared base classes to eliminate duplication. Prefer composition over copy-paste.+- **Encapsulation**: Use accessor methods for internal state (e.g. `set_current_task()`), not direct `_attribute` assignment from outside.+- **Provider-specific config**: Keep provider-specific fields (e.g. `nim_settings`) in provider constructors, not in the base `ProviderConfig`.+- **Dead code**: Remove unused code, legacy systems, and hardcoded values. Use settings/config instead of literals (e.g. `settings.provider_type` not `"nvidia_nim"`).+- **Performance**: Use list accumulation for strings (not `+=` in loops), cache env vars at init, prefer iterative over recursive when stack depth matters.+- **Platform-agnostic naming**: Use generic names (e.g. `PLATFORM_EDIT`) not platform-specific ones (e.g. `TELEGRAM_EDIT`) in shared code.+- **No type ignores**: Do not add `# type: ignore` or `# ty: ignore`. Fix the underlying type issue.+- **Backward compatibility**: When moving modules, add re-exports from old locations so existing imports keep working.+## COGNITIVE WORKFLOW1. ANALYZE: Read relevant files if you have not already. Do not guess.-2. PLAN: Use thinking mode to map out the logic. Identify the root cause or required changes.+2. PLAN: Use thinking mode to map out the logic. Identify the root cause or required changes. Order changes by dependency.3. EXECUTE: Fix the cause, not the symptom. Execute smartly and carefully4. VERIFY: Run tests or linting. Confirm the fix via logs or output.5. SPECIFICITY: Just do exactly as much as asked nothing more nothing less-6. PROPAGATION: Making changes has impacts across files so proapagate changes correctly+6. PROPAGATION: Making changes has impacts across files so propagate changes correctly## SUMMARY STANDARDS-- Summaries must be technical and granular.+- Summaries must be technical and granular.- Include: [Files Changed], [Logic Altered], [Verification Method], [Residual Risks].## TOOLS