Claude Code and Codex plugins their authors filed under "productivity". All categories.
Record, build, tailor, and review resumes in a single-column ATS-safe house style: keeps an honest experience record the user owns, generates a formatted Google Doc from that record, an interview, or an existing resume file, aims a copy at a specific job description, and audits any resume against a pre-submission checklist without ever inventing achievements.
3 skills
SlideTwin recreates a high-fidelity editable PowerPoint slide from one reference image with callable local reconstruction tools.
2 skills · 1 MCP server
Flow agentic workflow plugin with setup, update, and execution skills.
3 skills
Build your weekly routine, plan each day's focus blocks, and suggest tasks for you to review, with the local `rmd` CLI.
3 skills
Use Specflow workflows from Codex through the local specflow mcp server.
1 skill · 1 MCP server
Ad hoc web UI for multi-select choices with drag-and-drop reordering.
1 skill · 1 MCP server
Herdres operator skill.
2 skills
Routes each prompt to the Agent Skills that fit (or none) with a local hybrid retriever and a calibrated gate model.
1 skill · hooks
Lead with the big idea, put the one next action right under it, and say it terse. Substance never compressed.
1 skill · hooks
Switch back and forth between Claude Code, Codex, and beyond — searchable access to every coding-agent session on this machine.
1 skill · 1 MCP server
File-driven project management for AI agents — every repo change becomes one tracked unit in flat YAML, driven by a CLI + MCP. Setup bootstraps the full stack (npm or straight from GitHub).
1 skill · 1 MCP server
Automatic persistent memory that saves preferences, corrections, and project context. Works invisibly in the background.
1 skill · hooks
Typed, calibrated judgments from TypeSafe Jev: classify, score, check, a batched multi-question ask, a many-item triage with server-side file reads, and model discovery.
1 skill · 1 MCP server
SNARC — salience-gated memory for Claude Code. Scores every tool use on 5 dimensions (Surprise, Novelty, Arousal, Reward, Conflict) at capture time. Only stores what's salient. Automatic context injection, confidence decay, dream-cycle pattern extraction.
1 MCP server · hooks
Codex-native ocul-pm journal and planner guidance
1 skill · hooks
Self-amendment MCP server framework with layered skill architecture (L0/L1/L2) and blockchain-backed auditability. Skills can evolve, promote, and audit themselves through 29+ MCP tools. Prerequisites: Ruby 3.0+ and 'gem install kairos-chain' required for MCP server functionality.
1 skill · hooks
Process information for real understanding with the explain-back skill: explain from memory, grade against the real source, struggle-then-teach per gap, output Mochi cards and a resumable session record. An antidote to the illusion of clarity.
1 skill · 1 MCP server
Meta-skills that observe the Claude Code session itself: Stop-hook gated learning-candidate recording, session-pointer breadcrumbs into auto-memory
1 skill · hooks
Extract and analyze large PDFs (3MB-10MB+) with minimal token usage. Preserves 100% of content while achieving 12-103x token reduction through local extraction, semantic chunking, and intelligent caching. Perfect for technical documentation (NIST, ISO, AWS guides), compliance frameworks, and research papers. One-time extraction, instant querying.
1 skill · 1 agent
Extract and analyze large Excel workbooks (1MB-50MB+) with minimal token usage. Lossless extraction of all formulas, cell formatting, and complex table structures while achieving 20-100x token reduction through local extraction, semantic chunking, and intelligent caching. Perfect for compliance matrices, financial models, security audit logs, and data tables. Tested with 287K cells and 15K formulas.
1 skill · 1 agent
Extract and analyze Word documents (1MB-50MB+) with minimal token usage. Lossless extraction of all text, tables, formatting, and document structure while achieving 10-50x token reduction through local extraction, semantic chunking by headings, and intelligent caching. Perfect for policy documents, contracts, technical reports, and any large Word document.
1 skill · 1 agent
X/Twitter topic intelligence — scrapes real tweets, engagement metrics, and KOL accounts for content research
2 skills
Enforces the gws-multi-account layout: blocks bare `gws` CLI commands that do not set GOOGLE_WORKSPACE_CLI_CONFIG_DIR, and ships the matching skill so Claude picks the right account automatically.
1 skill · hooks
Create distinctive HTML one-pagers with curated aesthetics, converts to PDF. Anti-AI-slop design.
1 skill · hooks
Turns a "should we?" into a memo-ready recommendation grounded in expert consensus.
1 skill · 1 MCP server
Pressure-tests a plan into realistic timelines and delivery risks.
1 skill · 1 MCP server
Safely inspect and control physical computers through one or more JetKVM devices.
1 skill · 1 MCP server
LLMLingua-2 prompt compression hook for Claude Code — cut token usage by ~55%
1 skill · hooks
A report inbox for AI agents. Publish self-contained HTML reports, read them back as context, and share a stable link with the team.
1 skill · 1 MCP server
Use bounded Jev evidence selection to keep broad local search results out of Codex context.
1 skill · 1 MCP server