Added Prompt for tools #68
Reference in New Issue
Block a user
Delete Branch "64-prompt-for-tools"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
src/app/api/tools/plan_memory_tool.pysrc/app/api/tools/instructionsAvrei voluto testare meglio ma Google ha i modelli saturi oggi e provando con QWEN3 come team leader non ottengo risultati, è stupido e si perde malissimo.
Pull Request Overview
This PR refactors toolkit instruction strings by externalizing them from inline code to dedicated markdown files. It also relocates the
PlanMemoryToolto theapi/toolsdirectory and significantly enhances agent prompt documentation with detailed tool usage guidance, comprehensive examples, and improved workflow patterns.CryptoSymbolsToolsintegration to team leader workflow for cryptocurrency name resolutionReviewed Changes
Copilot reviewed 16 out of 16 changed files in this pull request and generated 4 comments.
Show a summary per file
_load_instructions()method to load instructions from external markdown file_load_instructions()method to externalize social toolkit instructions_load_instructions()method to externalize plan memory instructions_load_instructions()method to externalize news toolkit instructions_load_instructions()method to externalize market toolkit instructionsComments suppressed due to low confidence (4)
src/app/api/tools/plan_memory_tool.py:23
_load_instructions()method lacks error handling for missing instruction files. If the markdown file doesn't exist or is unreadable, this will raise an unhandled exception during toolkit initialization. Consider adding a try-except block with a fallback to a default instruction string or a more informative error message.src/app/api/tools/instructions/social_instructions.md:1
src/app/api/tools/instructions/news_instructions.md:1
src/app/api/tools/instructions/market_instructions.md:1
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
The
_load_instructions()method lacks error handling for missing instruction files. If the markdown file doesn't exist or is unreadable, this will raise an unhandled exception during toolkit initialization. Consider adding a try-except block with a fallback to a default instruction string or a more informative error message.The
_load_instructions()method lacks error handling for missing instruction files. If the markdown file doesn't exist or is unreadable, this will raise an unhandled exception during toolkit initialization. Consider adding a try-except block with a fallback to a default instruction string or a more informative error message.The
_load_instructions()method lacks error handling for missing instruction files. If the markdown file doesn't exist or is unreadable, this will raise an unhandled exception during toolkit initialization. Consider adding a try-except block with a fallback to a default instruction string or a more informative error message.The
_load_instructions()method lacks error handling for missing instruction files. If the markdown file doesn't exist or is unreadable, this will raise an unhandled exception during toolkit initialization. Consider adding a try-except block with a fallback to a default instruction string or a more informative error message.It is as expected since the missing or corrupted instruction must be a blocking error
Esiste già un metodo simile in
__init__.pydentro agents/prompt.Bastava richiamare quello o crearne uno universale da mettere da qualche parte, non replicarlo per ogni tool.
Esiste già un metodo simile in
__init__.pydentro agents/prompt.Bastava richiamare quello o crearne uno universale da mettere da qualche parte, non replicarlo per ogni tool.
@@ -0,0 +8,4 @@name: strstatus: Literal["pending", "completed", "failed"]result: str | NoneEsiste già un metodo simile in
__init__.pydentro agents/prompt.Bastava richiamare quello o crearne uno universale da mettere da qualche parte, non replicarlo per ogni tool.
Esiste già un metodo simile in
__init__.pydentro agents/prompt.Bastava richiamare quello o crearne uno universale da mettere da qualche parte, non replicarlo per ogni tool.
Esiste già un metodo simile in
__init__.pydentro agents/prompt.Bastava richiamare quello o crearne uno universale da mettere da qualche parte, non replicarlo per ogni tool.