Merge pull request #31 from huggingface/agent-improvements

Fix premature agent loop exit in long-running autonomous tasks

feat: add headless CLI mode with local filesystem tools and rich terminal rendering

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

feat: LLM-based pruning for complex tool outputs (hf_jobs)

fix: also skip plan_tool from deterministic pruning

fix: skip research tool outputs in deterministic pruning

Add deterministic pruning of old tool outputs before LLM compaction

refactor: simplify tool call validation and make interrupts cancel tool execution

fix: correct context_length init and emit tool_call events for malformed calls

fix: handle interrupt during streaming tool calls without corrupting context

fix: recover malformed tool calls instead of crashing session

fix: normalize dict tool_calls from litellm streaming to proper ToolCall objects

feat: upgrade CLI with local tools, slash commands, and interrupt support

fix: protect system prompt in undo, scan backwards for dangling tool calls, gate input on SDK status

fix: patch dangling tool calls on error, move undo to context manager

fix: improve compaction summary prompt, remove sidebar status bar

fix: route all HF token usage through session.hf_token from OAuth

feat: restore sandbox tools and proactive compaction

feat: merge HF Space improvements

v3 prompt and tool desc rework

adding compaction

frontend update

adding whoami to system prompt

integrated hf skills into the prompts

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

Added tools to create repos, view them, upload files to them (private). added a clock tool.

added the explore subagent, quality ok but not perf

Fixed assistant -> assistant messages after compaction

prompt update and loading from file

prompt change

added compaction every x tokens

intermediary commit

working events and session

minimal impl. of the agent