Categories › AI & Agents › Coding-agent workflow (specs, rules, CLAUDE.md, commands, hooks)
Software design skills for AI coding agents, inspired by A Philosophy of Software Design.
Remote server: your requests and data go to clairvoyance.fyi. Claude Code: claude mcp add --transport http clairvoyance https://clairvoyance.fyi/mcp · Claude app: Settings → Connectors → Add custom connector.
The tools this server listed when we connected to it, in its own words.
deep-modules — Measures module depth: whether the interface is simple relative to the implementation behind it. Use when an interface has too many parametemodule-boundaries — Evaluates where module boundaries are drawn and whether modules should be merged or split. Use when deciding whether to combine or separate information-hiding — Checks for information leakage across module boundaries, including temporal decomposition and false encapsulation. Use when modules change tpull-complexity-down — Checks whether complexity is pushed to callers or absorbed by implementations — the direction complexity flows. Use when callers must do siggeneral-vs-special — Evaluates whether interfaces are appropriately general-purpose. Use when checking interface generality, when if-branches or parameters serveerror-design — Reviews error handling and exception design, applying the "define errors out of existence" principle. Use when reviewing error handling, wheabstraction-quality — Evaluates whether abstractions provide a genuinely different way of thinking or are structurally shallow. Use when adjacent layers feel redunaming-obviousness — Reviews naming quality and code obviousness via the isolation test, scope-length principle, and consistency audit. Use when names feel vaguecomments-docs — Reviews comment quality and documentation practices: the four comment types, comments-first workflow, and comment rot. Use when reviewing costrategic-mindset — Assesses whether code reflects strategic or tactical thinking, including the 10-20% investment rule and tactical-tornado patterns. Use when design-it-twice — Generates and compares at least two fundamentally different design alternatives on concrete criteria before committing. Use when the user ascode-evolution — Evaluates whether changes to existing code maintain or degrade design quality. Use when the user asks to review a diff or PR (for example becomplexity-recognition — Diagnoses whether complexity exists and where it comes from, using the three-symptom, two-root-cause framework. Use when code feels harder tred-flags — Scans code against 17 design smells (the book's 14 named Red Flags plus 3 process-stage signals) and produces a structured diagnostic reportdesign-review — Orchestrates a structured design review, running the other skills as a diagnostic funnel from complexity triage to a full red-flags sweep. Udiagnose — Routes a vague symptom or complaint to the most relevant Clairvoyance skill via a decision tree. Use when someone describes a problem but dofetch-reference — Fetch a supporting file from a Clairvoyance skill's references/ folder, when the skill's instructions point to one. Available: information-hNot tested by KitScout yet. When our harness runs it, pros, cons and the tasks we tried appear here, with the date and evidence.