safeagents adds --log-only mode: observe agent calls without blocking, with live dashboard
yb2698 · x · 2026-10-29
Not ready to let it block anything? safeagents ships a --log-only mode: every tool call is still checked, but the model's response is forwarded untouched, so the agent behaves exactly as before.
- Live dashboard: a terminal panel shows each call's verdict — FLAG for calls it would have blocked, WARN for milder atif-scan findings. A screenshot shows one run with 94% PASS and 6% BLOCK across 124 calls.
- Audit trail: redirect output to a file (safeagents --upstream … --log-only > calls.log) for one plain line per call (time, verdict, tool, command, reason); flagged calls are kept in /.safeagents/calls.jsonl with full arguments, atif-scan findings and the task.
Related event: safeagents adds pluggable decision models and log-only mode(2 posts)→
More from coding & agent
- qwen-code PR Sizes All Qwen4 Models at 1M-Token Context Window — haider1 · 2026-10-29
- AI Engineer: Make Custom Evals a First-Class Citizen, Not Bigger Models — paulabartabajo_ · 2026-10-29
- Running 55 Claude Code subagents in one workflow for batch search tasks — AlchainHust · 2026-10-29
- Free DataTalks.Club x Nebius course: production AI agents for generative media — Al_Grigor · 2026-10-29
- MCP server makes Claude/ChatGPT cite datasheet pages for every part value — Winter-Drop-1334 · 2026-10-29
- Two devs used Claude to build a browser-based PS2 emulator, TabStationToo — yousriii · 2026-10-29