Skip to content

docs: scope llms-full.txt fetch to local harnesses - #106

Merged
Jammy2211 merged 1 commit into
mainfrom
feature/llms-txt-census-fixes
Jul 31, 2026
Merged

docs: scope llms-full.txt fetch to local harnesses#106
Jammy2211 merged 1 commit into
mainfrom
feature/llms-txt-census-fixes

Conversation

@Jammy2211

Copy link
Copy Markdown
Collaborator

Summary

AGENTS.md told every harness to check the autolens_workspace catalogue (llms-full.txt) before writing a script from scratch. That file is ~30k+ tokens and grows linearly with script count — a connector chat that ingests it pays the cost on every subsequent turn (the census's explanation for sluggish ChatGPT sessions). Local agents still grep llms-full.txt; chats now route from the compact llms.txt and read only the specific script they need.

Part of the llms.txt census (PyAutoLabs/autolens_workspace#451).

Generated by the PyAutoLabs agent workflow.

🤖 Generated with Claude Code

https://claude.ai/code/session_01GGPARVAg5oUT7rWRDMTA4E

…e#451)

The workspace catalogue is ~30k+ tokens and grows with script count; a
connector chat that ingests it pays the cost on every subsequent turn. Local
agents grep llms-full.txt; chats route from the compact llms.txt and read only
the specific script.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GGPARVAg5oUT7rWRDMTA4E
@Jammy2211
Jammy2211 merged commit b9c10a9 into main Jul 31, 2026
2 checks passed
@Jammy2211
Jammy2211 deleted the feature/llms-txt-census-fixes branch July 31, 2026 23:58
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.

1 participant