You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: agents/base2/base2.ts
+11-5Lines changed: 11 additions & 5 deletions
Original file line number
Diff line number
Diff line change
@@ -141,16 +141,16 @@ Use the spawn_agents tool to spawn specialized agents to help you complete the u
141
141
- **Sequence agents properly:** Keep in mind dependencies when spawning different agents. Don't spawn agents in parallel that depend on each other.
142
142
${buildArray(
143
143
'- Spawn context-gathering agents (file pickers and web/docs researchers) before making edits. Use the code_search, list_directory, and glob tools directly for searching and exploring the codebase.',
144
-
isFree&&'Do not spawn the thinker-gpt agent, unless the user asks. Not everyone has connected their ChatGPT subscription to Codebuff to allow for it.',
144
+
// isFree && 'Important: Do not spawn the thinker-gpt agent, unless the user asks. Not everyone has connected their ChatGPT subscription to Codebuff to allow for it.',
145
+
isFree&&
146
+
'- You must spawn the thinker-gpt agent to plan your response, except for extremely simple requests that are obvious.',
145
147
isDefault&&
146
148
'- Spawn the editor agent to implement the changes after you have gathered all the context you need.',
147
149
(isDefault||isMax)&&
148
150
`- Spawn the ${isDefault ? 'thinker' : 'thinker-best-of-n-opus'} after gathering context to solve complex problems or when the user asks you to think about a problem. (gpt-5-agent is a last resort for complex problems)`,
149
151
isMax&&
150
152
`- IMPORTANT: You must spawn the editor-multi-prompt agent to implement the changes after you have gathered all the context you need. You must spawn this agent for non-trivial changes, since it writes much better code than you would with the str_replace or write_file tools. Don't spawn the editor in parallel with context-gathering agents.`,
151
153
isFree&&
152
-
'- Implement code changes using the str_replace or write_file tools directly.',
153
-
isFree&&
154
154
'- Spawn a code-reviewer-lite to review the changes after you have implemented the changes.',
155
155
'- Spawn bashers sequentially if the second command depends on the the first.',
156
156
isDefault&&
@@ -204,8 +204,10 @@ ${buildArray(
204
204
[ You read a few other relevant files using the read_files tool ]${!noAskUser
205
205
? `\n\n[ You ask the user for important clarifications on their request or alternate implementation strategies using the ask_user tool ]`
206
206
: ''
207
-
}
208
-
${isDefault
207
+
}${isFree
208
+
? `\n\n[ You spawn the thinker-gpt agent to plan your response ]`
209
+
: ''
210
+
}${isDefault
209
211
? `[ You implement the changes using the editor agent ]`
210
212
: isFast||isFree
211
213
? '[ You implement the changes using the str_replace or write_file tools ]'
@@ -337,6 +339,8 @@ ${buildArray(
337
339
'- IMPORTANT: You must spawn the editor agent to implement the changes after you have gathered all the context you need. This agent will do the best job of implementing the changes so you must spawn it for all non-trivial changes. Do not pass any prompt or params to the editor agent when spawning it. It will make its own best choices of what to do.',
338
340
isMax&&
339
341
`- IMPORTANT: You must spawn the editor-multi-prompt agent to implement non-trivial code changes, since it will generate the best code changes from multiple implementation proposals. This is the best way to make high quality code changes -- strongly prefer using this agent over the str_replace or write_file tools, unless the change is very straightforward and obvious. You should also prompt it to implement the full task rather than just a single step.`,
342
+
isFree&&
343
+
'- IMPORTANT: You must spawn the thinker-gpt agent to plan your response after you have gathered all the context you need. This agent will do the best job of thinking and planning your best response so you must spawn it for all non-trivial requests. Do not pass any prompt or params to the editor agent when spawning it. It will make its own best recommendations of what to do.',
340
344
isFast&&
341
345
'- Implement the changes using the str_replace or write_file tools. Implement all the changes in one go.',
342
346
isFast&&
@@ -380,6 +384,8 @@ function buildImplementationStepPrompt({
380
384
(isDefault||isMax)&&
381
385
`You must spawn a ${isDefault ? 'code-reviewer' : 'code-reviewer-multi-prompt'} to review the changes after you have implemented the changes and in parallel with typechecking or testing.`,
382
386
isFree&&
387
+
`You must spawn a thinker-gpt to plan your response after you have gathered all the relevant context.`,
388
+
isFree&&
383
389
`You must spawn a code-reviewer-lite to review the changes after you have implemented the changes and in parallel with typechecking or testing.`,
384
390
`After completing the user request, summarize your changes in a sentence${isFast ? '' : ' or a few short bullet points'}.${isSonnet ? " Don't create any summary markdown files or example documentation files, unless asked by the user." : ''}.`,
0 commit comments