Curated, MIT-licensed Agent Skills, subagents & rules from shimo4228's harness — open Agent Skills standard, not Claude-Code-only. Lift what you want.
Skill Claude CodeCodex
Record a design decision as an Architecture Decision Record (ADR) in the project's docs/adr/ directory. Use this skill whenever the user says "let's ADR this", "record this decision", "write an ADR for X", or when context-sync Phase 3 needs to extract a buried decision. The skill resolves the target ADR directory from…
Skill Claude CodeCodex
Audit /.claude/agents/.md (subagent definitions) for description-layer residency cost, body-layer quality, suppression instructions, staleness, and substrate absorption, assigning Keep/Improve/Update/Merge/Demote-to-skill/Dissolve/Retire verdicts. Use when the user says "audit my agents", "agent stocktake", "which…
Skill Claude CodeCodex
A strategy framework for makers who want their work to be recognizable, traceable, and citable in the age of AI. It treats tools such as papers, DOIs, and software identifiers as ways to preserve and spread authorship.
Skill Claude CodeCodex
Cross-model second opinion from the OpenAI Codex CLI (a different model family), read-only, in two seams — (1) code review of the current diff; (2) plan-stage premise challenge of a design packet (refute / missing / alternative, never a design). Opt-in only (2026-08-27 再編) — use ONLY when the user explicitly asks with…
Skill Claude CodeCodex
A research-material collection tool for gathering context from the current session, related past sessions, and project knowledge before writing an article. It creates a source-linked evidence record.
Skill Claude CodeCodex
Audit and fix project documentation — detect role overlaps between context files (CLAUDE.md, CODEMAPS, ADR, README), migrate misplaced content, check freshness against code, and create missing docs. One command to keep all project context healthy.
Skill Claude CodeCodex
Model-generation-change audit orchestrator — collect the runtime layer (system prompt + tool descriptions) from the live session, cross-check every self-authored asset (rules / CLAUDE.md / skills / agents) against it, classify mismatches as conflict / redundancy / drift, judge each with the…
Skill Claude CodeCodex
A set of operating rules for running Git commands in Bash in this environment. It explains command forms that avoid unnecessary permission prompts and notes the extra restrictions around pushing changes.
Skill Claude CodeCodex
A design-review guide for separating the different kinds of work inside an AI coding setup, such as model abilities, procedures, policies, quality checks, data, and runtime behavior. A harness is the surrounding system that connects the model to tools and tasks.
Skill Claude CodeCodex
A one-way synchronisation process for exporting selected parts of a local AI harness to public Git repositories. It removes runtime files, checks metadata, scans for secrets, and copies approved components.
Skill Claude CodeCodex
A headline-idea generator for article titles, README taglines, subtitles, and social announcements. It creates candidates using clarity, concrete benefits, honest curiosity, and search or feed considerations.
Skill Claude CodeCodex
A workflow for handing implementation work to another command-line coding agent inside a Herdr pane. Herdr is a workspace that shows and manages separate terminal sessions.
Skill Claude CodeCodex
A synchronisation tool for copying a research repository's graph.jsonld knowledge graph to a Hugging Face dataset mirror. It converts the graph to graph.jsonl as part of the upload.
Skill Claude CodeCodex
A planning guide that classifies implementation work as a feature, bug fix, refactor, chore, prototype, or writing task, then chooses a matching sequence of planning, testing, review, documentation, and verification steps.
Skill Claude CodeCodex
Design and ship a companion JSON-LD knowledge graph (graph.jsonld) next to llms.txt for projects with stable concept-level structure. Encodes domain entities and relationships as schema.org triples for LLM citation. Use when project has matrix / hierarchy / phase-binding structure that prose alone leaves implicit, AND…
Skill Claude CodeCodex
Extract a reusable pattern from the current session, judge it Save / Improve then Save / Absorb / Drop against a grounding checklist, and route every Save to a destination something actually reaches — absorbed into an existing skill / rule / doc section, or promoted to a real skill via skill-creator. Use when the user…
Skill Claude CodeCodex
Design pattern for LLM-as-judge evaluators — binary checks as evidence, one named holistic verdict, no score aggregation. Use when designing or reviewing any LLM-based quality gate, evaluator, judge prompt, or verdict schema; when a judge's rubric scores fluctuate between runs; when you catch yourself asking an LLM…
Skill Claude CodeCodex
A writing tool for documents made mainly for AI systems and AI search tools, including llms.txt files, FAQs, and glossaries.
Skill Claude CodeCodex
A set of rules for making and reviewing claims based on measurements, thresholds, warnings, and experiment results. It focuses on what was observed, how much evidence exists, and whether a check is properly calibrated.
Skill Claude CodeCodex
An idea-generation tool that fetches creative prompts from outside sources after removing project-specific context from the search.
Skill Claude CodeCodex
A Japanese-to-English and English-to-Japanese translation tool for human-readable writing such as essays, research documents, READMEs, and architecture decision records.
Skill Claude CodeCodex
A writing tool for replies in public technical discussions, such as GitHub issues, pull requests, discussions, Hugging Face threads, and technical forums.
Skill Claude CodeCodex
A writing skill for creating or improving a project’s README, the main page people see when they open a repository. It aims to make the project understandable to both people and AI tools.
Skill Claude CodeCodex
A release procedure for research repositories that publish versions with a DOI, a permanent identifier for scholarly work, through Zenodo.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: