1
00:00:00,000 --> 00:00:01,282
Welcome to the show, everyone!

2
00:00:01,322 --> 00:00:05,835
I'm Lachlan Reed, here with James Turner,
and James, if you've ever had your

3
00:00:05,888 --> 00:00:11,752
workflow ground to a halt by enterprise
security red tape, you'll appreciate this.

4
00:00:12,785 --> 00:00:19,111
Claude Code version 2.1.158 just dropped,
and it finally brings Auto Mode to

5
00:00:19,211 --> 00:00:24,950
Bedrock, Vertex AI, and Palantir Foundry
for Opus 4.7 and 4.8.

6
00:00:26,000 --> 00:00:27,760
Oh, that is massive!

7
00:00:28,148 --> 00:00:32,007
So developers stuck behind those corporate
cloud gateways because of strict data

8
00:00:32,065 --> 00:00:36,085
compliance don't have to suffer through
"approval fatigue" anymore.

9
00:00:36,165 --> 00:00:39,441
They can actually use the agent's full
autonomous loop.

10
00:00:40,000 --> 00:00:40,877
Exactly.

11
00:00:40,957 --> 00:00:42,552
No more babysitting the terminal.

12
00:00:42,659 --> 00:00:43,993
To kick this off, you just set

13
00:00:44,113 --> 00:09:29,673
`CLAUDE_CODE_ENABLE_AUTO_MODE=1`

14
00:09:29,673 --> 00:09:29,673
in your environment.

15
00:09:29,673 --> 00:09:36,154
Once that's live, the agent can execute
complete write-run-correct cycles entirely

16
00:09:36,181 --> 00:09:42,078
on its own-no manual keypress approvals
needed for every single file change or shell

17
00:09:42,150 --> 00:09:43,038
execution.

18
00:09:44,000 --> 00:09:48,883
Right, but giving an agent free rein to
execute shell commands and modify code

19
00:09:48,953 --> 00:09:51,448
autonomously inside an enterprise VPC...

20
00:09:52,530 --> 00:09:54,329
that's going to make security teams sweat.

21
00:09:54,469 --> 00:09:56,163
How are they handling the paper trail?

22
00:09:57,000 --> 00:09:58,197
They actually thought of that.

23
00:09:58,341 --> 00:10:03,401
They've introduced a new OpenTelemetry
flag: `OTEL_LOG_TOOL_DETAILS=1`.

24
00:10:03,401 --> 00:10:08,596
When you enable that, it captures the
exact tool parameters and inputs the agent

25
00:10:08,676 --> 00:10:09,156
uses.

26
00:10:09,369 --> 00:10:13,794
It's a complete audit trail of what the
agent tried to do and when.

27
00:10:14,000 --> 00:10:19,197
Okay, so security gets their deep
telemetry, and developers get their speed.

28
00:10:19,902 --> 00:10:21,158
That's a fair trade.

29
00:10:21,268 --> 00:10:24,472
And hey, did they fix that annoying Git
lock issue?

30
00:10:25,000 --> 00:10:25,720
They did!

31
00:10:26,490 --> 00:10:30,203
They finally fixed the bug that was
leaving git worktrees locked.

32
00:10:30,463 --> 00:10:35,155
Now, when the agent is done, your
workspace cleanup is actually clean.

33
00:10:35,368 --> 00:10:37,794
No more manual unlocking faff.

34
00:10:38,000 --> 00:10:39,039
Thank goodness.

35
00:10:39,183 --> 00:10:44,562
This really feels like Claude Code is
maturing from a cool local CLI toy into

36
00:10:44,586 --> 00:10:48,082
something you can actually run at scale in
a highly regulated environment.

37
00:10:48,182 --> 00:10:49,997
Powered by Jellypod AI.

38
00:10:51,000 --> 00:10:51,799
Too right, mate.

39
00:10:51,919 --> 00:10:53,964
Let's see how enterprise teams adopt it.

40
00:10:54,124 --> 00:10:55,484
See you all next time!