Hermes on AiSlackin

MoACubed: teaching Hermes which profiles actually work

Sun, 06 Sep 2026 00:00:00 +0000

I have been running Hermes with multiple profiles for a while. That gives me useful specialization, but it also creates an uncomfortable question:

Which profile should handle a given kind of task?

It is easy to answer that question with vibes. One profile feels sharper. Another writes cleaner code. A third catches edge cases the others miss. Then one impressive run becomes a permanent routing rule.

That is not evaluation. That is anecdote with a config file.

Argus: the intel agent sheds its cron plumbing

Thu, 27 Aug 2026 00:00:00 +0000

When I built personal-intel-agent, it did one thing: watch a topic, fold duplicate coverage into a single sourced card, and ping only when something materially moved. I later added a second engine for typed metric and price monitoring.

It now has a better name—Argus—and a cleaner boundary with Hermes Agent 0.21.

The original version had to carry some of its own scheduler plumbing because cron jobs were effectively stateless. Hermes 0.21 changed that. Cron jobs can now use persistent agent memory, carry the previous run’s output forward, skip the LLM when a deterministic source has not changed, keep a bounded per-job notepad, select reasoning effort per job, and deliver directly into a bot’s canonical chat.

hermes-shortcut-cheat-sheet: the shortcut list from source, not docs

Wed, 19 Aug 2026 00:00:00 +0000

The published Hermes Desktop docs list about twelve keyboard shortcuts. The real list lives in the app’s keybind registry — rebindable actions, fixed keys, and generated profile/session slot bindings — and nobody had transcribed it. This repo is that transcription, generated from source rather than written by hand, so it’s complete and it’s correct.

The output ships three ways: an interactive dark-themed HTML with per-OS columns (Mac / Windows-Linux), a light single-page 3-column print layout for hanging on a wall, and PNG renders of both. A PROMPT.md contains the exact prompt to regenerate everything from a newer version of the source — paste it into any coding agent session and the sheet stays correct across app releases.

hermes-skills: a catalog for skills that survive machine rebuilds

Wed, 19 Aug 2026 00:00:00 +0000

Hermes Agent skills are just folders with a SKILL.md, which is exactly why they rot: they end up scattered across profiles and machines, unversioned, unrecorded. This repo is the fix I built for myself — a single maintained collection where every skill is registered in a catalog, installable with one command, and documented well enough that future-me (or anyone) can tell what it does without reverse-engineering the prompt.

The structure is deliberately boring: a _template scaffold you copy, a skills/README.md catalog you register into, and an installer script. Stateful skills keep tracker archives and metric snapshots on the host (under ~/.intel/ for Argus), never in the repo; durable user preferences now use Hermes’s native persistent memory.

hmem: benchmarking the memory providers fighting for your agent's long-term recall

Tue, 04 Aug 2026 00:00:00 +0000

Every agent framework eventually hits the same wall: the context window ends and the conversation forgets. The memory-provider ecosystem that sprang up to fix this is large, fast-moving, and mostly unmeasured — a lot of READMEs, few numbers. hmem is my attempt to bring order: a living comparison of fifteen providers compatible with Hermes Agent, from 62K-star mem0 down to single-file SQLite curiosities.

The part that separates it from a link list is the benchmark discipline. Providers only get scores when a real measured adapter exists — Hindsight and Mnemosyne are listed as “not measured” precisely because their simulated stubs were excluded. The current controlled run compares BM25 lexical baseline (0.367 correctness) against Hermes built-in memory (0.333) over 30 reviewed scenarios in 10 categories across three isolated runs: 180 result documents, zero failures, machine-readable JSON published alongside the writeup.

hermes-sendme-file-transfer: teaching an AI agent to move files peer-to-peer

Thu, 30 Jul 2026 00:00:00 +0000

Cloud file transfer means accounts, size caps, and someone else’s server reading your bytes. SendMe — built on iroh — does direct peer-to-peer transfer with a ticket you hand to the receiver, falling back to a relay only when NAT punching fails. This skill teaches Hermes Agent to drive all of it correctly.

The skill covers the parts that actually bite in practice: the long-running sender lifecycle (the process must stay alive while transfers complete), safe handling of bearer-like tickets (whoever holds the ticket can fetch — treat it like a credential), completion verification rather than trusting exit codes, and the Android/Termux quirks when the receiving end is a phone.

quill-of-hermes: LAN OCR that eats PDFs and multi-page TIFFs

Thu, 30 Jul 2026 00:00:00 +0000

OCR as a service usually means cloud APIs, per-page billing, and your documents on someone else’s disk. This is the local counterargument: a FastAPI wrapper around OvisOCR2 running GGUF via llama.cpp’s llama-mtmd-cli, serving your LAN. POST an image, a PDF, or a multi-page TIFF; get back clean Markdown.

The design decisions worth noting: each page is sent to the model independently and results are joined with a horizontal rule, so a 50-page PDF degrades gracefully instead of failing whole. Jobs are serialized to keep the GPU sane, there’s a five-minute timeout per page so one bad scan can’t wedge the queue, and temp files are cleaned unconditionally. The service deliberately stays separate from llama-swap because multimodal llama-server support for this model wasn’t proven — llama-mtmd-cli was.

karen_strikes_back: a five-persona review pod that tears your repo apart

Mon, 20 Jul 2026 00:00:00 +0000

Code review by one model is one opinion. Code review by five models with different personalities, priorities, and axes to grind finds things no single pass catches. That’s the premise of karen_strikes_back: a fleet of five Hermes Agent profiles — the name is the joke, Karen is the orchestrator and her workers are the ones who actually strike.

Each worker is a distinct lens with its own SOUL.md persona and pinned model config. critic is blunt and evidence-led, citing file:line for every defect — SQLi, missing authz, broken idempotency. bsneng reviews like a fresh Masters grad who’s read every Nielsen heuristic and will tell you which one you violated. poweruser is the insufferable PC enthusiast who checks whether every knob is exposed and will block shipping over telemetry-on-by-default. struggler is the self-deprecating advocate for users who bounce off jargon, hidden steps, and “just know” flows. Karen fans the repo out to all four in parallel via delegate_task, runs her own heuristic eval, then dedupes and severity-ranks everything into one combined report.

hermes-proxmox-multihost: one skill, N standalone Proxmox servers

Fri, 17 Jul 2026 00:00:00 +0000

Most home-lab Proxmox tooling assumes one cluster. Reality for a lot of us is several standalone servers, accumulated over the years, each with its own login, its own quirks, and no interest in joining anybody’s cluster. This skill — built on danielbitpro/hermes-proxmox-control, extended with a host registry and a multi-host helper — lets one Hermes Agent session manage all of them.

The part I’m proudest of is the failure model: get_all() wraps each connection in its own try/except, so when pve3 is down for maintenance, listing VMs across the fleet returns results from pve1 and pve2 plus a clean error line for pve3 — not a crash. Secrets resolve through Bitwarden Secrets Manager by UUID at call time; the registry on disk maps hostnames to secret IDs and never stores a credential itself.

Giving Hermes an Efficient Soul

Thu, 25 Jun 2026 00:00:00 +0000

I have been spending a lot of time with Hermes Agent lately. If you do not know what that is, short version: it is an agentic AI framework that runs locally, connects to whatever LLM backend you point it at, and operates autonomously on tasks you define. Think of it as the connective tissue between you, your local models, your cloud providers, and everything else you want an AI to touch.