This gist contains the Hermes Agent dog skill.
Install by copying SKILL.md into a Hermes skill directory, for example:
mkdir -p ~/.hermes/skills/productivity/dog
curl -L <raw-SKILL.md-url> -o ~/.hermes/skills/productivity/dog/SKILL.md
mkdir -p ~/.hermes/skills/productivity/dog/references
| name | hermes |
|---|---|
| description | Send commands, questions, and work requests to a user-configured remote Hermes ACP agent over SSH, then return the complete acpx/Hermes output without summarizing. Use this skill whenever the user asks to talk to, ask, delegate to, or run work through their remote Hermes agent or always-on remote agent. Configure hostnames, session names, paths, and behavior with environment variables such as $REMOTE_HERMES_HOST and $REMOTE_HERMES_SESSION_NAME. |
| compatibility | Requires network access to $REMOTE_HERMES_HOST, SSH authentication for that host, Node/npm with npx or a global acpx install, and a working Hermes ACP command on the remote host. |
Use this skill to communicate with a user-configured remote Hermes agent through ACP over SSH.
Alex from Ramp Labs describes an experiment in making Ramp Sheets more self-maintaining by connecting background coding agents to production observability. The core shift is that AI has reduced the effort of writing code, so the bottleneck moves to maintenance, monitoring, triage, and directing agents at the right problems. Ramp's answer is not merely "an agent that codes", but a system where agents inspect live behaviour, create monitors, respond to alerts, tune noisy signals, and generate pull requests for real issues.
The talk begins with Ramp Labs' broader AI work, including Ramp Sheets, Latent Briefing, and steering vectors, but quickly centres on a practical problem: once AI can handle much of the coding, engineers spend more time deciding what needs attention.
"code maintenance"
Eric Zakariasson argues that agentic software work is moving from pair-programming toward managed production systems: developers increasingly define intent, constraints, feedback loops, and verification rather than writing every line themselves. Cursor is not fully at the "software factory" stage yet, but parts of its product and company already run with high autonomy, offering practical lessons for teams trying to build similar systems.
The talk frames AI coding progress as a climb through levels of autonomy:
In this AI Engineer talk, Ryan Lopopolo argues that modern coding agents change the scarce resource in software engineering. The bottleneck is no longer typing implementation code, but designing the systems, context, guardrails, review loops, and operating practices that let agents reliably do the full job. His core claim is that engineering teams should reorganise around delegation, systems thinking, context management, and automation of acceptance criteria, rather than treating agents as autocomplete or code-writing assistants.
"Code is free."
Lopopolo frames coding agents as a major shift in engineering economics: implementation is becoming abundant, while human time, human/model attention, and model context window remain scarce. Engineers now have access to the equivalent of many parallel implementers, limited mai
| tmp() { | |
| mkdir -p ~/.tmp | |
| # Cleanup directories older than 23 days | |
| find ~/.tmp -mindepth 1 -depth -Btime +23d -delete 2>/dev/null | |
| local timestamp=$(date +%Y-%m-%d-%H-%M-%S) | |
| local dir | |
| if [[ -n "$1" ]]; then |
| <instructions> | |
| Create a structured summary of <input />, scaled to approximately [num-pages] page(s) when rendered. | |
| ## Structure | |
| ### Title | |
| Start with a `##` heading that captures what is being summarised (e.g., the article title, report name, or a descriptive title you create). | |
| ### Opening Paragraph | |
| Immediately after the title, write a paragraph stating the core thesis or takeaway. No heading for this paragraph — just the prose. |
A userscript that opens the current page in archive.ph with a single click or keyboard shortcut.
Handy for finding an archived copy of a paywalled or blocked web page.
Use this skill to create or edit images using OpenAI's GPT Image 1.5 model.
Unzip gpt-image-1-5.zip into ~/.claude/skills or ~/.skillz/ or wherever you keep skills locally. Or upload it to the Claude app.