Agent Skills by Jonathan Cimadevilla
Showing 27 vetted skills indexed across 1 GitHub repositories.
tel
Route whitelisted third-party service actions through the Trusted Execution Layer.
morning
Consolidates GitHub PRs, assigned issues, Brew updates, and dotfiles activity into a standup digest.
Grey-Area Arsenal
Track and archive controversial GitHub repositories by category with scheduled scans and JSON artifacts.
openhuman-bridge
Map OpenHuman Jarvis patterns into TEL scheduling and tool-output compression.
design
Generate HTML/CSS/SVG mockups from DESIGN.md and run Lighthouse audits.
consolidate-memory
Consolidate Claude session logs into tiered memory files.
checkpoint
Write markdown snapshots and HANDOFF.json artifacts to preserve session progress.
nonstop
Arms a Stop-hook to keep agents running through multi-step tasks without terminating the turn.
route
Route unattended AI tasks between local iMac agents and Claude Code Routines.
isa
Generate a verifiable Ideal State Artifact with pass/fail criteria and YAML frontmatter.
wired-up
Block the Stop action when orphan source files have zero references.
Arsenal
Consolidate system tools, routing rules, and operational commands into a capability index.
health
Check iMac development environment health and report broken items with repair commands.
pulse
Compress generated code while preserving identical behavior.
onboard
Discover repository structure and classify project shape for AI-OS documentation.
recall
Search MemPalace for prior verbatim decisions and conversation context.
skill-creator
Create Claude skills with SKILL.md YAML frontmatter under ~/.claude/skills/.
autobrowse
Iteratively run browser traces and update navigation strategies until tasks pass.
audit
Analyze repositories for quality and risk issues, then output a prioritized markdown punch list.
research-scout
Discover and validate new research and workflow information from web communities.
forge
Provide branching form and screen recording React components with backend callbacks.
autonomous-loop
Install project memory files and loop-launch setup for autonomous self-improvement cycles.
website-design-stack
Generate a production rubric for six-section landing pages with tier-based animation and stack selection.
kimi-webbridge
Automate authenticated browser navigation and interaction via the Kimi WebBridge daemon.