Skill Claude CodeCodex
Broad-to-narrow research Orchestration - start with a wide census, harden the scope from returned Evidence, and specialize Subagents only once the exact situation is proven. TRIGGER when the Architect says "progressive research", "start broad", or when dispatching research or debugging Subagents into a situation whose…
Skill Claude CodeCodex
Mandatory contract for every proposing-state turn. The classifier names it when the Architect types /propose. Routes every decision to its layer, researches until the Proposal names files, methods, and data changes, orders the Decisions by dependency, shows each one with its code and options, and fixes the findable…
Skill Claude CodeCodex
Produce the Evidence that proves work is done before any completion claim. TRIGGER before reporting a Task complete, writing report.md, or claiming fixed, passing, or pre-existing. DO NOT TRIGGER while still implementing.
Skill Claude CodeCodex
Check a diff for Capability regressions — loss of User-facing Capability ("the User can no longer X") or loss of system Capability ("our system can no longer Y"). Maps changed code to affected Critical Path and system Capabilities, traces each through the code, and reports any that no longer work. TRIGGER on…
Skill Claude CodeCodex
Source-grounded external research for current library, API, framework, service, and vendor behavior. TRIGGER when the Task asks for current facts, docs, releases, compatibility, adoption, or external technical Decisions. DO NOT TRIGGER for in-codebase Architecture tracing; use /trace or the explorer agent.
Skill Claude CodeCodex
Read the changeset yourself, review it with the code-reviewer and ponytail Agents on the same target at once, triage their findings with /triage, and iterate until a round returns no breaking finding. TRIGGER on "review the changes", "code review", on the Review before a changeset is presented to the Architect, and…
Skill Claude CodeCodex
Collaborative Shaping with the Architect: capture requirements, boundaries, shapes, fit checks, research unknown mechanics, and transition to Modeling then Slicing. TRIGGER when the Architect asks to shape, scope, compare approaches, or turn a rough idea into a modeled change.
Skill Claude Code
Build and publish the page the Architect reviews an Architecture Proposal on — one self-contained page toggling between the start state and the final state across architecture diagrams, annotated file trees, and database changes. TRIGGER when the Architect fires this Command by name. DO NOT TRIGGER to show…
Skill Claude CodeCodex
Draw every code change in the reply as a diff, an annotated file tree, or a class diagram, and keep ideas in prose. TRIGGER on every reply that explains code, Architecture, a change, a name, or a finding, and whenever the Architect says "show me". DO NOT TRIGGER to score a set of options; that is /rank.
Skill Claude CodeCodex
Turns a shaped and modeled change into vertical Slices with acceptance criteria and Verification requirements. Each Slice ends in demo-able functionality. The Plan becomes an immutable contract.
Skill Claude CodeCodex
Turn returned reviewer findings into one graded report grouped by who each problem affects, and act on the gate. TRIGGER after review Subagents return findings — from /review, /regressions, /user-testing, or any audit — and before results reach the Architect. DO NOT TRIGGER while reviewers are still running, or for a…
Skill Claude CodeCodex
Deep research of the systems behind a change — name the systems, show the architecture, read the git history for intent, trace the code, attack the understanding, and report in a layered file another Agent can act on coherently. TRIGGER when a change lands in a system the Agent has not mapped, when the Architect says…
Skill Claude CodeCodex
Tests code changes by tracing real Critical Paths. Lists Critical Paths affected by uncommitted changes, spawns one Subagent per Critical Path to trace Execution and find gaps, then evaluates Architecturally. TRIGGER when the Architect says "user test", "test the Critical Paths", or "trace Critical Paths".
Skill Claude CodeCodex
Verify the whole changeset the way the User meets it: the suite, the Review gate, the Critical Paths, once every Task has landed. TRIGGER when the last Slice is staged, when an autonomous run reaches functional completion, or on "verify the changes". DO NOT TRIGGER with Tasks still open, where the Orchestrator judges…
Skill Claude CodeCodex
Builds a Voice.md profile from a writer's existing samples — reverse-engineers the patterns that make their copy sound like them, then writes a structured, reusable profile the rest of the team applies at write time. TRIGGER when asked to capture, define, or learn a voice, build or regenerate a Voice.md, analyze…
Skill Claude CodeCodex
Find the BIG IDEA of a piece — the one fresh, charged idea the whole thing hangs on, spiky enough that the reader stops and feels it. TRIGGER when a piece needs its core idea found or sharpened, when a draft reads as flat or generic, or when a headline or angle has no single idea underneath it. DO NOT TRIGGER for line…
Skill Claude CodeCodex
Read one web page for research — a plain-HTTP fetch that passes bot walls headless browsers cannot, cleaned to its main content, cached, and recorded in the source registry. TRIGGER when a research or review step needs a page's text. DO NOT TRIGGER to screenshot a page or drive an interactive surface (screenshot…
Skill Claude CodeCodex
Read the piece AS the product's owner-approved buyer — believable, valuable, would they pay. Belief lives here. TRIGGER when a draft needs the buyer's reaction, a believability read, or when the Buyer Reviewer loads it for the review-as-a-buyer stage. DO NOT TRIGGER for whether a claim is technically backed…
Skill Codex
The sole de-slop pass — AI tells, metaphors and metaphorical verbs, synthetic cadence, template smell, and the additive distinctiveness check. Anything whose cause is genericness. TRIGGER when de-slopping a draft, hunting AI tells or generic SaaS phrasing, or when the editor loads it for the check-for-ai-writing…
Skill Codex
Edit drafts into sharper, more human writing while preserving the writer's personal voice, or detect AI-slop patterns without rewriting. Use when the user wants a draft clearer, more direct, more opinionated, or less AI-sounding, or asks whether writing reads as AI.
Skill Claude CodeCodex
Trace every claim in a piece to a research record or a product fact — proof presence lives here. Unsourced claims are listed, never shipped, never auto-killed; they escalate to the chief. TRIGGER when fact-checking a draft, verifying claims and proof, or when the Fact Checker loads it for the check-the-claims stage.…
Skill Claude CodeCodex
Attack whether the market, buyer, problem, and language a research or strategy artifact claims actually EXIST — the reality-attack pass. Not improvement, destruction of fabricated reality. Every item gets a 1-to-100 existence-confidence rating with reasoning, default low; internal coherence is not evidence. TRIGGER on…
Skill Claude CodeCodex
Judge whether the piece's argument holds — right buyer at the right awareness and sophistication, sections answering the buyer's real questions in a real order, and nothing starring that shouldn't lead. A strategic judgment gate argued from the buyer's perspective, not a checklist. Findings block. TRIGGER on the…
Skill Claude CodeCodex
The between-sections pass — coherence, the seams between parts, the skim layer (the title chain alone tells the story), structural repetition, and set-level continuity across a multi-piece deliverable. Returns findings and suggested edits; the writer folds them in. Never mutates the draft. TRIGGER when checking a…
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: