Skill Claude CodeCodex
Emit and resume cross-session handoff artifacts. Emit captures a session's mission, state, and next steps into a versioned file at explicit checkpoints; resume boots a new session from the artifact behind a continue/revise/archive decision checkpoint. Use for context handoffs between sessions, operators, or their…
Skill Claude CodeCodex
Triage raw human-channel input pasted directly into the session (chat exports, email threads, PR comments, meeting notes) — or dropped as files in a project inbox — into typed, deduplicated, decision-ready items with numbered recommendations. Strictly selection-gated — nothing dispatches without an explicit operator…
Skill Claude CodeCodex
Convert an incident into a durable agent-policy amendment. Use after any workflow failure, regression discovery, lost-work incident, or repeated agent mistake — researches root cause, finds the instruction gap, proposes the minimal amendment, and applies it only on explicit approval.
Skill Claude CodeCodex
Stage-1 acceptance verification — requirements-alignment check between the active plan/spec and the actual implementation diff, producing a GAP ledger bound to a commit digest. Use before accepting, PR-ing, or handoff-packaging any implementation work. Empirical QA and human-facing acceptance reports are later stages…
Skill Claude CodeCodex
L2 mechanics for type in motion: variable-font axis animation, kinetic text, per-letter effects, scroll- and pointer-linked type, and typographic state transitions. Fires on implementing or auditing type-specific motion, or when chained from a higher layer. Ownership of motion budgets, timing doctrine, and…
Skill Claude CodeCodex
L0 prose mechanics for the typography stack: drafting, rewriting, and critiquing text for cadence, punctuation discipline, clarity, voice, and medium fit. Fires on explicit user requests to draft, rewrite, or critique prose for cadence, punctuation, clarity, voice, or a named medium profile, and when another skill…
Skill Claude CodeCodex
Engineering floors for typesetting static HTML/CSS: semantic structure, measure, leading, fluid clamp() scales, variable-font axes, font loading, accessibility, i18n. Use when implementing or auditing static web typography, or when chained by a deliverable skill that needs a type-engineering layer beneath its pins.…
Skill Claude CodeCodex
Disciplined disk-cleanup workflow for reclaiming storage without destroying live runtimes, unpushed work, or unrecoverable data. Use for ANY disk-space cleanup, storage reclamation, "what can we delete", purge/prune campaigns, or when asked to free up space — survey, classify, verify, capture, delete, report.
Command Claude Code
Write an episode checkpoint receipt for the current session's episode.
Command Claude Code
Run the full codified workflow — Design → Debate → Plan → Review → Implementation → Review → Closeout — for a task, without re-describing the stages.
Skill Claude CodeCodex
Standalone, idempotent episode closeout. Mandatory-referenced by every terminal path (Atlas post-F-wave, in-session completion, lite completion). Inputs: episode manifest + linked artifacts + receipts ONLY (bounded sessionevidence refs: message IDs + immutable excerpts; never full transcripts). Outputs: problem/intent…
Skill OpenCode
OMO/opencode adapter for the feature rollout protocol (docs/rollout-protocol.md + scripts/rollout-monitor.py). Use when deploying or enabling ANY new harness capability — a daemon, watcher, pipeline stage, or agent write-path — to arm the monitoring ladder durably, wire the evidence sources, and execute gates. Covers…
Skill Claude CodeCodex
Atlas-level review orchestration handler. Manages the complete review workflow — delegates review tasks, processes findings, enforces cycle limits, and prevents loops. Loaded by the orchestrator when REVIEW-ENFORCER fires.
Skill Claude CodeCodex
OS-level computer use on the local X11 desktop via the cua-driver daemon: screenshots, AT-SPI element trees, background (co-work) mouse/keyboard input, and window management for NATIVE applications only. Use for native GUI apps without APIs, OS dialogs, and desktop automation; NOT for any browser/web task (use…
Skill Claude CodeCodex
Structured adversarial analysis protocol. Quick single-agent modes (challenge, panel, pre-mortem, red team) plus a decision-review protocol where one agent proposes, one critiques, the proposer revises, and a binding/advisory judge panel decides ADOPT/REVISE/REJECT/ESCALATE.
Skill Claude CodeCodex
Infrastructure and deployment helper for server setup, port management, Docker/docker-compose, local services, and deployment automation. Maintains port registry to avoid conflicts.
Skill Claude CodeCodex
Emit and resume cross-session handoff artifacts. Emit captures a session's mission, state, and next steps into a versioned file at explicit checkpoints; resume boots a new session from the artifact behind a continue/revise/archive decision checkpoint. Use for context handoffs between sessions, operators, or their…
Skill Claude CodeCodex
Triage raw human-channel input pasted directly into the session (chat exports, email threads, PR comments, meeting notes) — or dropped as files in a project inbox — into typed, deduplicated, decision-ready items with numbered recommendations. Strictly selection-gated — nothing dispatches without an explicit operator…
Skill Claude CodeCodex
Safely merge a worktree branch into main with strict local state tracking, lock discipline, and deterministic rollback.
Skill Claude CodeCodex
Skill "parallel-dev.tmp" from EZotoff/ez-omo-config, covering parallel-dev, should i parallelize? decision framework, mandatory preconditions (all must be yes), risk assessment (proceed if acceptable) and decision matrix.
Skill Claude CodeCodex
Patch the live OpenCode binary with a minimal fix. Use when fixing bugs in OpenCode's Go/TypeScript binary — checkout the exact live version, apply the fix, build, install, and test. NEVER build from dev branch to patch a stable release.
Skill Claude CodeCodex
Convert an incident into a durable agent-policy amendment. Use after any workflow failure, regression discovery, lost-work incident, or repeated agent mistake — researches root cause, finds the instruction gap, proposes the minimal amendment, and applies it only on explicit approval.
Skill Claude CodeCodex
Reader-first writing for any deliverable a human reads to understand a result: HTML reports, MD design docs, debate-result summaries, survey dashboards, executive briefs, handoff notes, panel-review syntheses. Writes for a reader who has NOT read the preceding material. NOT for: bug reports, incident reports, status…
Skill Claude CodeCodex
Register a new retryable error pattern in the error retry registry. Use when encountering errors that should trigger automatic retry.
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: