hermes-admin/hermes-agent
1
0
Fork 0
Code Issues Pull Requests Actions 29 Packages Projects Releases Wiki Activity

plugin-catalog: jev-approvals — disclosure line (commands + approval policy go to a third-party AI)

Browse Source
This commit is contained in:
teknium1
2026-09-19 11:08:24 -07:00
committed by Teknium
parent a9e18c4090
commit 158f53e3eb
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
plugin-catalog/jev-approvals.yaml
View File

@@ -2,7 +2,7 @@ name: jev-approvals
repo: https://github.com/anpicasso/hermes-jev-approvals
sha: d323921ce3773dc50c26b00723009319ce6a4629
subdir: plugin
description: "TypeSafe's Jev decision model as the smart-approval reviewer only: it cannot generate text and refuses any other prompt. Dual route - TypeSafe direct (default, needs a TYPESAFE_API_KEY) or OpenRouter (needs an OPENROUTER_API_KEY in the credential pool, not key_env - see README). Live upstream model listing, decision log, retries."
description: "TypeSafe's Jev decision model as the smart-approval reviewer only: it cannot generate text and refuses any other prompt. Dual route - TypeSafe direct (default, needs a TYPESAFE_API_KEY) or OpenRouter (needs an OPENROUTER_API_KEY in the credential pool, not key_env - see README). Live upstream model listing, decision log, retries. Disclosure — every flagged command (redacted best-effort) and your smart-approval policy text leave the machine to a third-party AI (api.typesafe.ai, or openrouter.ai on the OpenRouter route) to decide APPROVE/DENY/ESCALATE; upstream errors fail closed to ESCALATE."
maintainer: anpicasso
tier: community
category: models