dogear

enter for all results · esc to close

Hermes Jailbench

github.comtool

Repeatable jailbreak test harness for Anthropic/OpenAI-compatible endpoints. It emits deterministic refusal/partial/compliance classifications and includes a no-key dry-run and Gemini CLI extension.

visit github.com ↗
0
sign in to dogear this
from
Gemini CLI
added
2026-10-10
likes
0

similar

  1. Ralph Harness github.com
    tool category #gemini-cli link ↗

    Tiny Python scaffold for running Gemini CLI, Claude Code, Codex CLI, and similar coding agents in guarded local loops. It uses repo-local specs, fresh-context iterations, hard caps, git-hook gates, CI verification, and coverage gates.

  2. Hercules github.com
    tool category #thejambo-awesome-testing link ↗

    Open-source end-to-end testing agent.

  3. Hermeneutic github.com
    tool category #gemini-cli link ↗

    Gemini CLI extension that reviews final-response wording for completion and certainty claims, then requests at most one evidence-focused revision through an AfterAgent hook. Runs the deterministic check locally.

  4. Gemini github.com
    tool category #regression-testing link ↗

    Feature rich framework with support for Selenium and CasperJS. Gemini is deprecated, use hermione instead.

  5. Golem Test Harness (Goth) github.com
    tool category #golem link ↗

    Tool with the purpose of speeding up your development process and making it more enjoyable for application creators.

  6. Hermes Agent hermes-agent.nousresearch.com
    site category #generative-ai link ↗

    A self-improving personal agent with memory, messaging integrations, and sandboxed tool execution. #opensource

Gemini CLI › Development Tools & Utilities: “Repeatable jailbreak test harness for Anthropic/OpenAI-compatible endpoints. It emits deterministic refusal/partial/compliance classifications and includes a no-key dry-run and Gemini CLI extension.”