…l-commit-title-pr

* origin/main:
  feat: show available toolkits (#37)
  adding in ability to provide per repo hints (#32)
  Apply ruff and add to CI (#40)
  added some regex based checks for dangerous commands (#38)
  chore: Update publish github workflow to check package versions before publishing (#19)
  chore: upgrade ai-exchange dependency (#36)
  fix: resuming sessions (#35)
  feat: upgrade `ai-exchange` to version `0.8.3` and fix tests (#34)
  fix: export metadata.plugins export should have valid module (#30)
  fix (#24)
  link to vs code extension (#20)
  Enable cli options for plugin (#22)
  Modified the readme to be more friendly to new users (#16)
  chore: gitignore generated lockfile (#15)
  add prompts (#11)
  conditionally publish only when config changes (#9)

* added the entry point for plugin with cli group option

* main:
  fix: typo in exchange method `rewind` (#54)
  fix: remove unsafe pop of messages (#47)
  chore: Update LICENSE (#53)
  chore(docs): update is_dangerous_command method description (#48)
  refactor: improve safety rails speed and prompt (#45)
  feat: make goosehints jinja templated (#43)
  ci: enforce PR title follows conventional commit (#14)
  feat: show available toolkits (#37)
  adding in ability to provide per repo hints (#32)
  Apply ruff and add to CI (#40)
  added some regex based checks for dangerous commands (#38)
  chore: Update publish github workflow to check package versions before publishing (#19)
  chore: upgrade ai-exchange dependency (#36)
  fix: resuming sessions (#35)
  feat: upgrade `ai-exchange` to version `0.8.3` and fix tests (#34)
  fix: export metadata.plugins export should have valid module (#30)
  fix (#24)
  link to vs code extension (#20)
  Enable cli options for plugin (#22)
  Modified the readme to be more friendly to new users (#16)

* added the entry point for plugin with cli group option

* added the entry point for plugin with cli group option

* added the entry point for plugin with cli group option

Issue block#22: AI was removing retry and settings fields when it should
only remove the context field documented in validation-changes.md.

Added a scannable ⚠️ CRITICAL RULES section at the top with 5 clear rules:
1. Only change what's explicitly listed in validation-changes.md
2. Don't remove any content except documented removals
3. Don't add any content except documented additions
4. Don't modify any content except documented changes
5. Preserve all other content exactly as-is

These generalized rules apply to all sections (tables, field specs,
validation rules) and should prevent the AI from making unintended
changes while keeping the instructions concise and scannable.

Issue block#22 Root Cause: AI used diff format for table row removal, which
removed more lines than intended (retry and settings rows were removed
along with context row).

Comprehensive fix with multiple layers of protection:

1. Critical Rule #4: Explicit guidance on table updates
   - Use old_str/new_str (NOT diff format)
   - Include COMPLETE row to remove
   - Include 1-2 rows before/after for context
   - Verify exact match
   - Double-check not removing adjacent rows

2. Update Strategy Table: Clear mapping of change types to actions
   - Field removed row includes IMPORTANT note about str_replace

3. Streamlined structure:
   - More scannable with better organization
   - Cleaner task description
   - Verification section reminding to check for unintended changes
   - Guidelines emphasizing precision

This multi-layered approach provides:
- Explicit technique specification (tells AI exactly how)
- Repetition (same guidance in multiple places)
- Concrete examples (shows what 'surrounding rows' means)
- Verification (asks AI to double-check)

File paths remain correct (./output/validation-changes.md, etc.)