Skip to content

[agent_farm] explain context crunching (Run ID: codestoryai_sidecar_issue_2067_a92e2423) #2068

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: main
Choose a base branch
from

Conversation

theskcd
Copy link
Contributor

@theskcd theskcd commented Feb 17, 2025

agent_instance: codestoryai_sidecar_issue_2067_a92e2423 Tries to fix: #2067

📝 Documentation Addition: Created comprehensive explanation of context crunching mechanism

  • Added: Detailed documentation of the context crunching process, including the token threshold logic (60k for standard LLMs, 120k for custom LLMs) and action node compression
  • Explained: The flow from agent_tool_use endpoint to ToolUseAgent context crunching, including the "toggle fort reasoning" feature and O3 model usage for summarization
  • Purpose: Provides clarity on how the system manages large contexts to prevent token limit issues while maintaining instruction coherence

@Colttzdll
Copy link

fuck you

Copy link
Contributor

agentfarmx bot commented Mar 5, 2025

Fix: Implemented message moderation for chat and tool use

  • Added: A message_moderation module to filter inappropriate language in user messages.
  • Updated: agent_session_chat and agent_tool_use functions to incorporate the moderation system. Please review!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[agent_farm] explain context crunching
2 participants