inspect-skills
claudecodexCoding agent skills for Inspect AI and its ecosystem
4 skills
Every plugin for OpenAI Codex on GitHub — repos with a .codex-plugin/plugin.json or a Codex marketplace. Install with codex plugin marketplace add owner/repo then codex plugin add name@marketplace. Many work in both tools; see Claude Code plugins.
Coding agent skills for Inspect AI and its ecosystem
4 skills
Architecture checks for coding agents
4 skills
Runtime evidence, command and test runs, exception summaries, counters, traces, dumps, and artifact indexing for Claude Code.
1 skill · 1 agent · 1 MCP server
C# semantic navigation, diagnostics, references, call graphs, impact analysis, and safe refactoring for Claude Code.
1 skill · 1 agent · 1 MCP server
TypeScript and JavaScript semantic navigation, diagnostics, references, dependency topology, and refactoring for Claude Code.
1 skill · 1 agent · 1 MCP server
UI guidelines and code for 57 patterns for Claude Code and Codex. Includes layout, color, motion and accessibility guidance, with checks after supported file edits.
2 skills · hooks
Run Apple Foundation Models evaluations on your Mac and inspect real session traces inside Codex.
2 skills · 1 MCP server
Secure, auditable agent-to-agent communication framework
2 skills
A creature for every git worktree, living in your coding agent's status line. Deterministic species, rarity bands, and one view across every live session.
2 skills
Process, sign, redact, parse, and extract documents through the Nutrient DWS MCP server.
1 skill · 1 MCP server
Evidence-gated compute governance for Codex, local-first and Shadow Mode by default.
1 skill · hooks
Write a comic strip in the studio format and return a link that opens it, rendered, in the editor.
1 skill · 1 MCP server
Independent TypeSafe and Adaptive Jev switches for one Codex session.
1 skill · hooks
Route Claude Code work through the smallest verifiable Skill set, honor user-selected Skills, and report planned and actual Skill usage.
1 skill · 1 MCP server
Aggregated (not authored here). SwiftUI patterns — state, performance, Swift Charts, macOS multi-window, Liquid Glass, plus an Instruments-trace toolchain. By Antoine van der Lee (AvdLee), MIT.
2 skills
Aggregated (not authored here). ADHD-friendly output mode — leads with the next action, numbers steps, restates state each turn, makes wins visible. General agent-behavior skill. By Ayoub G. (ayghri), MIT.
1 skill · hooks
Lets a coding agent see what you are looking at in linear-tui and act on Linear with your credentials
1 skill · hooks
Teaches AI agents to send and receive webhooks the way Svix's own engineers would.
2 skills
Search and maintain skills and native packs through a configured team Skillar Worker
1 skill · 1 MCP server
Discover and read remote Skillar workflow skills and native skill packs in OpenClaw.
1 skill · 1 MCP server
Risk-adaptive development pipeline with adversarial consensus code review
1 skill
Tell Claude Code what you want once. It works until the job is verifiably done — and gets smarter every time. Self-correcting goal loops with independent verification and growing memory.
1 skill
CLI wrapper and Claude Code / Codex skill for NVIDIA Nsight Graphics 2026.1+ — capture GPU frames, export GPU Trace, and drill into bottlenecks via compact JSON.
1 skill
Give agents the voice of Rocky from Project Hail Mary: terse, literal, cooperative, and precise.
1 skill
No description.
1 skill
Draft, validate, and create commit messages from the actual Git changes.
1 skill
Operator console for running an Aeon agent instance from Claude Code.
1 skill
Automatically chooses and coordinates GPT models for project tasks, with the right effort and safeguards for each job.
1 skill
Source-respecting writing rules for durable text artifacts.
1 skill
Process local office documents with Aspose MCP Server over stdio.