Merge pull request #31 from huggingface/agent-improvements

Fix premature agent loop exit in long-running autonomous tasks

refactor: append dynamic session context after static system prompt for cache stability

fix: update system_prompt_v3.yaml (the actual active prompt) to use research tool

feat: add research sub-agent tool, slim down main agent system prompt

feat: add data audit section to system prompt

feat: add hf_papers tool for paper discovery and reading

feat: restore sandbox tools and proactive compaction

feat: merge HF Space improvements

rewording

v3 prompt and tool desc rework

sandbox working and agent can use it

sandbox start

frontend update

added code panel, tool approval, tool calls logging

added whooooshing to api spec

adding whoami to system prompt

Remove unused prompt files

integrated hf skills into the prompts

enhanced prompt v0

main merge

new system prompt and push to hub automatic checks

system prompt and github tool desc. update

loading date in system prompt, search code update (still not there)

private tool handles spaces better + prompt improvements for reliability

main agent search has better vibes

(partially done) system prompt tells to use research agent

improved search agent prompt and descriptions

Merge branch 'main' into explore-tool

added a api search tool and refined the hf docs search tools

prompt changes, more robust hf token sending, banner

re-order system instructions

added the explore subagent, quality ok but not perf

small prompt change, printing job url before stream instead of id

pass ToolErrors to agent, correct HF token handling, simplify uv job calls

prompt update and loading from file