Skip to content

5dive

v0.1.39 Feature

This release adds 2 notable features for engineering teams evaluating rollout.

✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agents ai automation shell claude-code cli
+4 more
codex gemini self-hosted systemd

Affected surfaces

auth breaking_upgrade

Summary

AI summary

Fixed silent DM drops and enabled provisioning of opencode telegram agents.

Full changelog

opencode + telegram: customer-ready (DIVE-45)

Two fixes that together unblock shipping opencode telegram agents to customers:

  1. access.json always seeded. install_channel_for_opencode_agent previously only wrote access.json when --telegram-allowed-users was passed. The normal "provision bot, pair via first DM" flow left no file, and the bridge's loadAccess() defaults to block-everything (allowFrom:[], dmPolicy:allowlist) → the bot silently dropped every DM. Now it always seeds: dmPolicy:pairing with no allowed-users (first DM yields a pairing code), allowlist when users are given. An existing policy is never overridden.

  2. install.sh stages telegram-opencode. The installer staged telegram-codex/-grok/-agy but not -opencode, so a customer box couldn't provision an opencode telegram agent at all. Added the staging block (mirrors agy); update.sh --upgrade heals existing boxes at the next nightly refresh.

Verified on a fresh Hetzner box: the smoke's new opencode-telegram regression row passes (access.json pre-seeded with dmPolicy:pairing; plugin staged).

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track 5dive

Get notified when new releases ship.

Sign up free

About 5dive

All releases →

Related context

Beta — feedback welcome: [email protected]