@dreb/coding-agent 2.60.0 → 2.60.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -319,7 +319,7 @@ A trusted root permits lazy loading for that existing directory and all of its d
319
319
 
320
320
  Replace the default system prompt with `.dreb/SYSTEM.md` (project) or `~/.dreb/agent/SYSTEM.md` (global). Append without replacing via `APPEND_SYSTEM.md`.
321
321
 
322
- For persistent instructions that apply only to one exact provider/model pair, use `modelSettings` in `settings.json` with a canonical key such as `openai-codex/gpt-5.6-sol`. Set `systemPrompt` to replace dreb's built-in prompt or `appendSystemPrompt` to append instructions. The same mechanism works for local/custom models registered in `models.json`, and the prompt updates immediately when the model changes. Explicit session replacements (`--system-prompt`, `SYSTEM.md`, or SDK hooks) take precedence. See [Model settings](docs/settings.md#modelsettings).
322
+ For persistent instructions that apply only to one exact provider/model pair, set `systemPrompt` to replace dreb's built-in prompt or `appendSystemPrompt` to append instructions. Put the field directly on a custom `models[]` entry or built-in `modelOverrides` entry in `models.json`, or use `modelSettings` in `settings.json` with a canonical key such as `openai-codex/gpt-5.6-sol`. Configure prompt behavior in only one file for a canonical model: dreb fails loudly instead of applying implicit precedence when both files declare it. Prompt changes apply when the model changes, and `/reload` picks up edits from either file. Explicit session replacements (`--system-prompt`, `SYSTEM.md`, or SDK hooks) take precedence over model replacement; model append instructions still follow the selected base. See [Custom models](docs/models.md#model-configuration) and [Model settings](docs/settings.md#modelsettings).
323
323
 
324
324
  ---
325
325
 
@@ -431,6 +431,7 @@ export declare class AgentSession {
431
431
  getFilteredSkills(): import("./skills.js").Skill[];
432
432
  /** @deprecated Use getFilteredSkills() instead */
433
433
  private _getFilteredSkills;
434
+ private _resolveModelPromptSettings;
434
435
  private _rebuildSystemPrompt;
435
436
  /**
436
437
  * Send a prompt to the agent.
@@ -563,7 +564,7 @@ export declare class AgentSession {
563
564
  * resolves identically to how createAgentSession seeds it at startup.
564
565
  */
565
566
  private _refreshThinkingDisplay;
566
- /** Reject malformed target-model prompt settings before a model switch mutates session state. */
567
+ /** Reject malformed or conflicting target-model prompt settings before mutating session state. */
567
568
  private _validateModelPromptSettings;
568
569
  /**
569
570
  * Cycle to next/previous model.