Skip to content

Update Readme.md - #1

Merged
Kunlun-Zhu merged 4 commits into
mainfrom
update_dataset
Mar 10, 2025
Merged

Update Readme.md#1
Kunlun-Zhu merged 4 commits into
mainfrom
update_dataset

Conversation

@CharlieDreemur

Copy link
Copy Markdown
Collaborator

add huggingface dataset

@Kunlun-Zhu
Kunlun-Zhu merged commit 223f51b into main Mar 10, 2025
realtmxi pushed a commit that referenced this pull request May 23, 2025
CarmiBanxe referenced this pull request in CarmiBanxe/OpenManus-RL Jul 15, 2026
- agent/persona.py: PromptTemplate (safe {var} render), built-in PERSONAS
  (assistant/concise/coder/analyst), resolve_system_prompt (explicit > persona),
  Guardrails (max_input_chars + opt-in deny_patterns). S-18: NO content censorship
  — deny-list empty by default (Legion is intentionally uncensored); guardrails are
  operational (input size / operator-configured patterns) only.
- AgentConfig: persona/system_prompt/max_input_chars/deny_patterns fields.
- LegionAgent: prepends system prompt; guardrails.check in _prepare (raises
  GuardrailError). Applies to chat + stream + tools paths.
- agent_server: persona/guardrails from env (LEGION_PERSONA/SYSTEM_PROMPT/MAX_INPUT/
  DENY); guardrail violation -> HTTP 400.
- Tests: 14 passed. Regression agent/server/sessions/stream-tools 23. Full 526.
- Verified live (concise persona): '2+2' -> '4' (no preamble) — system prompt shapes
  output e2e. Status doc roadmap #1-3 marked done (S20/S21/S22).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
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.

2 participants