romiluz13/cc10x

Agent Common provides the shared preamble for cc10x agents, including memory loading, contract formatting, and output rules. Use it when an agent needs the standard workflow and read-only domain glossary.

Architecture designs greenfield systems by mapping functionality flows, components, APIs, dependencies, and observability. Use it for new systems or major multi-component features; for existing code with shallow modules, use codebase-hygiene instead.

Building writes production code with test-first discipline using RED-GREEN-REFACTOR, vertical slicing, and safe refactoring. Use it when implementing features or fixes that need reliable tests and code generation patterns.

Cc10X Update safely upgrades cc10x while preserving local changes. Use it to check for new versions, pull upstream, rebuild the cache, and reapply patches.

Cc10x Guide answers questions about cc10x itself, including installation, configuration, workflows, memory, hooks, and troubleshooting. It is for explanation only; setup, updates, and work requests route to other skills.

Cc10x Router routes code requests into build, debug, review, plan, triage, or orient workflows. Use it when a request needs immediate orchestration, task execution, validation, and persistence.

Code Review runs adversarial diff reviews with confidence-scored findings and file:line evidence. It also supports receiving review by checking external feedback before agreeing or acting on it.

Codebase Design defines vocabulary for shaping modules with small interfaces and deep implementations. Use it when designing seams, improving testability, or deciding whether an abstraction is shallow or deep.

Codebase Hygiene finds semantic duplicates and shallow modules that spread complexity across a codebase. It is read-only and advisory, used before routing any real changes through BUILD with full gates.

Debugging uses a feedback-loop-first approach to reproduce bugs before forming hypotheses. It covers root-cause tracing, blast-radius checks, and a four-phase investigation process for agent-driven debugging.

Diff-Driven Docs updates documentation after a build, staged commit, or PR when code changes are not yet reflected. It uses diff impact rules to decide whether business, technical, audit, or glossary docs need updates.

Domain Modeling actively shapes a project's glossary and domain terms by challenging fuzzy language, testing edge cases, and updating CONTEXT.md as the model evolves. Use it when defining or refining concepts, and in read-only mode to apply the project's established vocabulary.

Exploration turns rough ideas into validated designs through collaborative interview, or creates throwaway spike code to answer one design question. It is used when an agent needs to clarify scope before planning or prototype a risky choice.

Frontend supports UI authoring and read-only critique. Use authoring for accessibility, responsive layouts, loading and error states, performance, and DESIGN.md work; use critique to score built UI against a rubric.

MCP CLI discovers and calls MCP server tools on demand for a single task. Use it for quick research or debugging when you do not want to mount a server permanently.

Memory & Handoff loads and persists workflow state under .cc10x/ so context survives compaction. It also exports a redacted handoff package for a coworker, another tool, or a fresh session.

Plan Review Gate checks saved plans and decision RFCs for feasibility, completeness, and alignment before execution. It is used after non-trivial planning when a fail-closed review should block unsafe or incomplete work.

Planning distills tasks into durable execution plans and decision RFCs. Use it to break work into bite-sized tasks, define validation, manage risk, and map functionality flow.

Research synthesizes web and GitHub findings into an answer with patterns, gotchas, and references. Use it when research files have already been gathered and you need a concise recommendation.

Resolving Merge Conflicts helps you work through an in-progress git merge or rebase conflict hunk by hunk. Use it when you need to inspect both sides, resolve by intent, run checks, and finish the merge or rebase.

Verification enforces proof-first validation for agent work. Use it to independently confirm a task actually achieved its goal, with evidence, exit codes, and scenario-level checks.