neuron-explorer
Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results
9 agent skill packs for Claude Code, Cursor, Codex and 70+ more agents: UX scenarios, gated delivery, multi-agent leases, skill authoring, design tokens, SEO/AEO audits, payment and agent-system patte
Developer toolsJavaScript · PythonGitHub @sshelegOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
Install and update nine Agent Skills packs for UX, delivery, design, integrations, SEO, agent systems, Telegram, coordination, and skill authoring.
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
A working method for coding agents, with a skill family and Project Observatory.
Route the task, preserve its intent, verify the result and leave a handoff another agent can use. The skill family supplies the instructions and checks; Project Observatory is a separately installed tool for observing projects and credential exposures. How the harness works.
Verify the resolved family: npx --yes sshlg-skills@latest list.
Detailed docs → · one page per skill,
every routing rule verbatim, and llms.txt
for the machines. · Follow @sshlg93 on XRead the full README ↗The full README is unavailable right now. Read it at the original link. The saved excerpt is shown below.
A working method for coding agents, with a skill family and Project Observatory.
Route the task, preserve its intent, verify the result and leave a handoff another agent can use. The skill family supplies the instructions and checks; Project Observatory is a separately installed tool for observing projects and credential exposures. How the harness works.
Verify the resolved family: npx --yes sshlg-skills@latest list.
Detailed docs → · one page per skill,
every routing rule verbatim, and llms.txt
for the machines. · Follow @sshlg93 on XRead the full README ↗