Skill Claude CodeCodex
Use when the user asks to shape fuzzy planning, product/design/engineering tradeoffs, scope/spec/ticket plans, or says "challenge this/help me think"; delegate to an advisor subagent. Good for rough ideas, second opinions, load-bearing decisions, or scope cuts. Skip scoped implementation, narrow bugs, reviews, and…
Skill Claude CodeCodex
Use when planning or delivering an LLM or agentic-system project, especially for experiment loops, walking skeletons, demos, data stories, scope control, and release-readiness evidence.
Skill Claude CodeCodex
Use when designing, implementing, reviewing, or debugging LLM-backed or agentic systems, including prompts, structured outputs, tool use, RAG, agent loops, orchestration, observability, security, cost/provider routing, and AI delivery plans.
Skill Claude CodeCodex
Use when evaluating prompts, LLM features, agents, RAG, judges, tool-use policies, or any stochastic behavior where one successful run is not reliable evidence.
Skill Claude CodeCodex
Use when setting up a project-local eval harness for prompts, agents, judges, RAG, tools, or other LLM behavior, especially when the default should be free/OSS promptfoo with repo-owned artifacts.
Skill Claude CodeCodex
Use when a pull request or merge request needs to be driven to merge — monitoring current-head checks, classifying failures, handling review threads, and merging only after repository-required gates, across GitHub, Forgejo, or GitLab.
Skill Claude CodeCodex
Use when preparing to edit a feature, bug fix, refactor, documentation or configuration update, packaging or release work, migration, operational change, or other substantive repository change; classify it before editing and explicitly decide all ten surfaces from repository or supplied evidence.
Skill Claude CodeCodex
Use when a pushed CI run fails or before later work or pushes; coordinate multiple agents through Tiber's CI-recovery owner/lease, reproduce the exact recovery record, permit only a causal repair or unchanged-SHA rerun, and require terminal success.
Skill Claude CodeCodex
Use when choosing whether and how verified work should be committed, pushed, reviewed, and delivered under a repository's local workflow policy; local-only mode does not authorize a commit unless the user or repository requires one.
Skill Claude CodeCodex
Use when a development request needs lifecycle routing for diagnosis, implementation, review feedback, PR or MR creation, readiness, CI failure, verification, or delivery.
Skill Claude CodeCodex
Use when completing or claiming readiness for local-only changes, a branch, pull request, merge request, or merge-to-main; final review applies even when repository policy forbids publishing, including scope growth, proposed work splits, and medium-risk review-budget checkpoints.
Skill Claude CodeCodex
Use when selecting a model and reasoning effort for a delegated task or escalating an agent assignment.
Skill Claude CodeCodex
Use when writing, reviewing, or proposing a Git commit message for authored development work, including after an implementation increment is complete.
Skill Claude CodeCodex
Use when receiving human, automated, GitHub, CodeRabbit, or other review feedback before changing code.
Skill Claude CodeCodex
Use when a bug, failing test, broken command, unexpected output, or confusing runtime behavior appears.
Skill Claude CodeCodex
Use when implementing a feature, bugfix, production behavior change, removal, or refactor to decide whether RED applies and, when it does, work from one failing public behavior test at a time.
Skill Claude CodeCodex
Use when claiming work is done, fixed, passing, ready, reviewed, committed, or safe to merge.
Skill Claude CodeCodex
Use when creating or editing skills in this marketplace or preparing skill behavior fixtures.
Skill Claude CodeCodex
Use when starting or making substantive changes to a project that should follow a strict, portfolio-grade engineering regime — the default standards for architecture, type-safety, error handling, testing, linting, ADRs, and review to apply as you work. (To set up the tooling that enforces them, use the scaffold…
Skill Claude CodeCodex
Use when setting up, bootstrapping, or scaffolding a new or existing project's tooling and config to enforce the engineering standards — generating a reproducible dev environment, pinned toolchain, strict lints, mutation and BDD harnesses, ADR structure, and CI/CD gates tailored to the detected stack.
Skill Claude CodeCodex
Use when a plugin, skill, prompt, tool, command, or agent behavior was surprising, wrong, partial, unsafe, brittle, or worth preserving as a future eval fixture; also use when the user asks to report, submit, capture, or file an eval case.
Skill Claude Code
Use when the user asks to add, capture, file, or record a new repository task or backlog ticket through Tiber from chat.
Skill Claude CodeCodex
Use whenever Tiber or origin/tiber is mentioned, including repository task tracking, EventCore publication conflicts, ambiguous pushes, tiber sync recovery, force-push safety, cross-worktree task-board coordination, setup, and task workflows.
Skill Claude CodeCodex
Use when making a repository worktree-ready for optional concurrent development, creating an explicitly requested linked worktree, or diagnosing worktree setup.
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: