ranade-oss widens the gate for its AI automation

A small workflow change makes it easier for scheduled GitHub automation to reach an AI agent with the ability to review and author pull requests.

securityworkflow

The fork now accepts a broader range of GitHub automation identities, rather than one exact bot account. It also explicitly admits scheduled runs, after those were apparently being rejected under the earlier setup.

The safeguards still matter: the agent is intended to operate under read-only permissions and without administrator-style access. But this is nevertheless a deliberate expansion of who can invoke a powerful automation path. The use of a name-based match, rather than a single exact identity, deserves scrutiny before anyone reuses this setup.

So what Legal-tech teams using AI-assisted engineering workflows should care because convenience in automated maintenance can quietly widen the security boundary.

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

2 commits from ranade-oss/ROSS-RanadeOSS, oldest first. Source extracted verbatim from the harvested git log.

SHA	Subject	Author	Date
`81de3deb`	Allow GitHub Actions bot Codex workflows	ranade-oss	2026-08-04	↗ GitHub
`c93bdfa5`	Fix scheduled Mike classifier authorization (#98)	ranade-oss	2026-08-09	↗ GitHub
* Allow scheduled Mike classifiers to run * Allow scheduled Mike classifiers to run

SHA

Subject

Author

Date

81de3deb

Allow GitHub Actions bot Codex workflows

ranade-oss

2026-08-04

↗ GitHub

c93bdfa5

Fix scheduled Mike classifier authorization (#98)

ranade-oss

2026-08-09

↗ GitHub

* Allow scheduled Mike classifiers to run

* Allow scheduled Mike classifiers to run

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-900.md from inside the repo you want the changes in.