Skip to content

Add doxa ingest --yolo: one-flag unattended agent fetching#14

Merged
0xadvait merged 1 commit into
mainfrom
yolo-mode
Jun 6, 2026
Merged

Add doxa ingest --yolo: one-flag unattended agent fetching#14
0xadvait merged 1 commit into
mainfrom
yolo-mode

Conversation

@0xadvait

@0xadvait 0xadvait commented Jun 6, 2026

Copy link
Copy Markdown
Owner

A first-class yolo mode on top of the launch-hardening opt-ins (which made agents safe-by-default). doxa ingest ... --yolo flips on the danger switches for that run only:

  • codex --dangerously-bypass-approvals-and-sandbox (= sources.codex.unsafe_bypass)
  • hermes --yolo (= sources.hermes.unsafe_yolo)
  • command shell (= sources.command.allow_shell)

So agent/command fetchers run fully unattended on trusted sources (a warning prints when engaged). Default stays safe.

doxa ingest https://target.example --via codex --yolo

+1 test (flag parses, opt-ins flip, agent argv honors them). Full suite 91 passed. README documents it.

🤖 Generated with Claude Code

Builds on the launch-hardening opt-ins (safe by default). `--yolo` flips on the
danger switches for a single ingest -- codex --dangerously-bypass-approvals-and-
sandbox (sources.codex.unsafe_bypass), hermes --yolo (sources.hermes.unsafe_yolo),
and the command shell (sources.command.allow_shell) -- so agent/command fetchers
run unattended on trusted sources. Prints a warning when engaged.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
@0xadvait
0xadvait merged commit a5fbabe into main Jun 6, 2026
@0xadvait
0xadvait deleted the yolo-mode branch June 6, 2026 19:50
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