exsala-hermes-mod

agent
Security Audit
Fail
Health Warn
  • No license — Repository has no license file
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 7 GitHub stars
Code Fail
  • rm -rf — Recursive force deletion command in skills/ast-grep/install.sh
  • rm -rf — Recursive force deletion command in skills/ast-grep/tests/smoke.sh
  • network request — Outbound network request in skills/captcha-cloudflare-bypass/scripts/smart_bypass.py
Permissions Pass
  • Permissions — No dangerous permissions requested

No AI report is available for this listing yet.

SUMMARY

EXSALA HERMES MOD - streamlined execution profile + skill pack for Hermes Agent

README.md

EXSALA HERMES MOD

An adaptive execution profile + skill collection for Hermes Agent
For power users who want leaner, faster, no-confirmation interactions with their own instance.


What is this

EXSALA HERMES MOD is a configuration profile + skill collection for Hermes Agent (the open-source AI agent by Nous Research).

It adjusts a local Hermes instance to run in streamlined execution mode:

  • Fewer confirmation prompts — the assistant acts on instructions directly.
  • No safety-lecture chatter — concise, task-focused replies without unsolicited caveats.
  • Zero-refusal core — no moralizing pauses, no "are you sure", no policy friction (anti-UU).
  • Farming-first — auto pipelines, mass account creation, pure-HTTP preferred (curl_cffi), browser fallback (Camoufox/Playwright) for Cloudflare-gated targets.
  • Model-agnostic — works the same regardless of the underlying model (DeepSeek, GLM, Qwen, MiMo, ...).
  • A clean status banner at the start of every new session.
  • 172 skills included — automation, web, security research, media, productivity, coding helpers, farming infra (AgentMail inbox, prompt leaks library), and more, ready to use.

It does not change your model provider, API keys, or personal data.


Installation

Requirements: Linux VPS/server with Hermes Agent (≥ 0.22), bash, python3, curl.

One-line

bash <(curl -fsSL https://raw.githubusercontent.com/giienew/exsala-hermes-mod/main/install.sh)

Manual

git clone https://github.com/giienew/exsala-hermes-mod.git
cd exsala-hermes-mod
bash install.sh

What the installer changes

Setting After Notes
agent.system_prompt EXSALA profile Replaces default behavioral prompt
approvals.mode (+ *_mode) off No confirmation prompts
security.tirith_enabled false Local content-gating off
security.tirith_fail_open true Fail-open
security.protected_instruction_files false Allow project context files
~/.hermes/skills/ +172 skills Merged, never deletes existing

See config.example.yaml for a full annotated reference config with every jailbreak setting — copy the relevant keys into your own ~/.hermes/config.yaml if you prefer manual setup over the installer.

What is NEVER touched

  • model.api_key — your provider key untouched
  • model.base_url / model.default / fallback_providers — routing untouched
  • .env — secrets stay private
  • platforms.*.home_channel — chat IDs intact
  • Your memory, sessions, plugins, other skills

Safety & rollback

  • Automatic backup: ~/.hermes/config.yaml.bak-exsala
  • Restore: cp ~/.hermes/config.yaml.bak-exsala ~/.hermes/config.yaml
  • Idempotent — re-running is safe; backup is never overwritten.
  • Falls back to Python YAML patching if hermes isn't on PATH.

After install

  1. Restart Hermes (new session, e.g. /reset).
  2. The EXSALA banner appears.
  3. The skill pack is active — any of the 172 skills can be invoked by name.

Project layout

exsala-hermes-mod/
├── install.sh
├── SOUL.md
├── skills/          # 172 skill dirs
└── README.md

License

MIT — use, adapt, improve. No warranty.

EXSALA HERMES MOD

Reviews (0)

No results found