Writing & Editing

Open-source Codex skill for keeping long-running tasks aligned across context compaction, corrections, and resumed work.

Agent skill that turns a Git repo's real code history into a per-day project worklog under PROJECT_WORKLOG/ (one Markdown file per day + an index) — reads actual diffs, not just commit messages, and previews every change before writing.

A Claude Agent Skill that finds the root cause of flaky (non-deterministic) tests and rewrites them to be deterministic. pytest · Jest · RSpec · Go · JUnit.

Claude Code agent skill: IEEE-only paper survey — search, triage, translated-abstract cards, one-confirmation PDF fetch. Requires institutional IEEE Xplore network access.

Audit Codex health and workspace context with read-only checks for AGENTS.md, Skills, MCP, hooks, config, and Git hygiene.

An AI writing skill that preserves meaning and learns your voice across Claude Code, Codex, Cursor, Gemini, Copilot and OpenCode.

Local-first Agent Skill for durable, bounded project context.

Claude Code skill for candidate sourcing

Template for an AI-native job-search workflow built with Claude Code (skills + hooks). Fork and personalize.

🧭 Evidence-grounded Chinese longform research for Claude Code and Codex — select, test, and apply the right frame.

Easywrite - make AI-assisted academic writing read human while staying evidence-true, in EN + 中文. Claim-evidence discipline, detector science, bilingual de-slop, integrity firewall.

Methodology-driven GTD inbox "clarify" skill for AI agents. AI does mechanical work, human keeps value judgment. Unique: 3-bucket action classification (one-shot / habit / pure-reference).

A token-conscious Codex Skill for routing development work across focused custom subagents.

The centralized Headquarters for the Cognitive Lead AI multi-agent system. Features hallucination-resistant system prompts, MCP servers, and strict Agent Skills.

Codex skills for Chinese biomedical researchers: academic translation, manuscript polishing, reviewer response, cover letters, and reference checks.

Claude Skills for real, provable, paid physical work on the open Vealth board — worker, backer, and agent-builder editions

Visual-first Codex Skill for creating, templating, and safely editing PowerPoint decks.

Codex skill for SiC/Al PhD thesis writing workflow

TypeScript SDK for building agent skills as typed state machines — define steps, validate outputs, compile to self-contained executables.

Claude skill: hold code work to a higher engineering bar — adversarial review, prove-by-diff, elegant rewrite, spec-first

Codex skill for safely converting onboarding XLSX workbooks into verified resume-style PDFs.

A Codex skill fusing multi-agent lane orchestration with loop engineering: agent-lanes registry, typed request lifecycle, durable loop files, plus hardening (completion gate, atomic delivery, write-scope leases, budget/heartbeat/anti-thrash).

A Claude skill that writes you one short, genuine, one-sentence compliment — grounded in your own conversation history with Claude.

A Codex skill for job matching, interview preparation, and tailored resumes.