cdeust

199 mods across 9 repositories, 109 stars between them.

ai-architect

145

cdeust/ai-architect-mcp-codebase

MCP server Claude CodeCodexCursor +2

MCP server "ai-architect" as configured in cdeust/ai-architect-mcp-codebase. Launched with ${CLAUDE_PLUGIN_ROOT}/bin/launch-plugin.sh.

4 2d ago A tokens not measured original MIT

impact-analysis

147

cdeust/ai-architect-mcp-codebase

Skill Claude CodeCodex

Trace the structural blast radius of a proposed or actual code change with ai-architect. Use for implementation planning, pull-request review, regression analysis, dependency checks, or identifying callers, importers, users, communities, and processes affected by a symbol or diff.

4 2d ago A 55 tokens original MIT

understand-codebase

148

cdeust/ai-architect-mcp-codebase

Skill Claude CodeCodex

Build an evidence-backed map of an unfamiliar repository with the ai-architect code graph. Use before implementing a feature, debugging an unfamiliar subsystem, explaining architecture, or making structural claims about symbols and dependencies.

4 2d ago A 45 tokens original MIT

validate-change-plan

149

cdeust/ai-architect-mcp-codebase

Skill Claude CodeCodex

Check whether a proposed implementation plan matches the current code graph. Use before coding or approving a plan that names files, symbols, dependencies, boundaries, or expected blast radius; this is structural plan validation, not full PRD or semantic correctness verification.

4 2d ago A 53 tokens original MIT

cdeust/ai-architect-mcp-codebase

MCP server Claude CodeCodexCursor +2

MCP server "ai-architect-mcp-codebase" as configured in cdeust/ai-architect-mcp-codebase. Launched with https://github.com/cdeust/ai-architect-mcp-codebase/releases/download/v0.11.1/ai.

4 2d ago A tokens not measured original MIT

ai-architect-mcp-spec

152

cdeust/ai-architect-mcp-spec

Plugin Claude Code

Stateless reducer that turns a feature description into a 9-file PRD. Multi-judge verification with weighted-average + Bayesian consensus, calibrated against externally-grounded oracles (schema/math/code/spec). Closed loop: per-judge Bayesian reliability calibration, Kaplan-Meier retry budgets, Clopper-Pearson KPI…

2 9d ago A tokens not measured original MIT

prd-gen

153

cdeust/ai-architect-mcp-spec

MCP server Claude CodeCodexCursor +2

MCP server "prd-gen" as configured in cdeust/ai-architect-mcp-spec. Launched with bash ${CLAUDE_PLUGIN_ROOT}/bin/ensure-deps.sh ${CLAUDE_PLUGIN_ROOT}. Needs 2 environment variables to run.

2 9d ago A tokens not measured original MIT

ai-architect-mcp-spec

156

cdeust/ai-architect-mcp-spec

Skill Claude CodeCodex

Action-driven PRD generation. The MCP server runs a stateless 20-step pipeline reducer (11 PRD-generation steps + 9 opt-in implementation steps behind a human gate); the host (Claude Code) executes each substantive action and feeds the result back via submitactionresult, looping until done. Multi-judge verification…

2 9d ago B 109 tokens original MIT

audit-prd

157

cdeust/ai-architect-mcp-spec

Skill Claude CodeCodex

Audit a multi-section PRD with deterministic Hard Output Rules and cross-section consistency checks. Use when reviewing a complete product requirements document, checking FR-to-AC coverage, acceptance-criteria numbering, story-point arithmetic, or test traceability.

2 9d ago A 51 tokens original MIT

validate-spec

158

cdeust/ai-architect-mcp-spec

Skill Claude CodeCodex

Validate one PRD or technical-specification section with deterministic Hard Output Rules. Use when checking a single overview, requirement set, user-story section, acceptance-criteria section, API specification, data model, security section, test plan, deployment plan, risk section, timeline, source-code section, or…

2 9d ago A 66 tokens original MIT

ai-architect-mcp-spec

159

cdeust/ai-architect-mcp-spec

MCP server Claude CodeCodexCursor +2

MCP server "ai-architect-mcp-spec" as configured in cdeust/ai-architect-mcp-spec. Launched with https://github.com/cdeust/ai-architect-mcp-spec/releases/download/v0.8.0/ai-arch.

2 9d ago A tokens not measured original MIT

session-optimizer

161

cdeust/session-optimizer

Plugin Claude Code

DEPRECATED meta plugin (v2.0.0): session-optimizer split into three independently installable plugins — context-guard (Stop-hook context budget + memory-writer checkpoint subagent + subagent spend tracker), refine-gate (UserPromptSubmit prompt-binding gate + /refine skill), and statusline (multi-line status bar +…

1 7d ago A tokens not measured original MIT

SessionStart

162

cdeust/session-optimizer

Hook

Runs when a session starts on startup, running an inline shell check. From cdeust/session-optimizer.

1 7d ago A tokens not measured original MIT

context-guard

163

cdeust/session-optimizer

Plugin Claude Code

Context-budget guard for long Claude Code sessions: a Stop hook enforces a per-model checkpoint protocol (mechanical stub + budgeted memory-writer subagent), and a SubagentStop tracker surfaces true session spend (main thread + subagents). Default protocol uses only vanilla Claude Code; a scoped memory layer is…

1 7d ago A tokens not measured original MIT

memory-writer

164

cdeust/session-optimizer

Agent

Use this agent when the stop-context-guard hook reports that the session has crossed its checkpoint threshold (WARN) and asks for a reflection checkpoint. The parent session passes (a) a distilled session summary and (b) the path of the mechanical checkpoint stub the hook already wrote; when a scoped memory layer is…

1 7d ago A 298 tokens original MIT

refine-gate

165

cdeust/session-optimizer

Plugin Claude Code

Prompt-binding gate for Claude Code: a UserPromptSubmit hook detects prompts carrying unbound references ("the SSE solution", "like before", "still broken") or ungrounded work requests, and injects the /refine procedure — build a binding table with evidence, separate symptom from goal, select an execution strategy…

1 7d ago A tokens not measured original MIT

refine

166

cdeust/session-optimizer

Skill Claude CodeCodex

Crossover layer between the user's intent and LLM execution: takes a raw prompt (terse, frustrated, ambiguous, or shorthand), resolves every vague reference to a concrete artifact, recalls past decisions and lessons, and compiles a verifiable execution contract BEFORE any code is touched. Invoke explicitly as $refine…

1 7d ago A 118 tokens original MIT

statusline

167

cdeust/session-optimizer

Plugin Claude Code

Multi-line Claude Code statusline with a discrete heat-track context bar tied to per-model checkpoint thresholds, one deduplicated cost ledger covering subagent spend, per-session telemetry (tok/s, compactions, prompt-cache countdown), rate-limit gauges with burn-rate pacing, and terminal-width fitting. Words, not…

1 7d ago A tokens not measured original MIT

statusline

168

cdeust/session-optimizer

Skill Claude CodeCodex

Install or update the session-optimizer statusline for Claude Code. Use when the user asks to "install the statusline", "set up the status bar", "configure the statusline", "update the statusline", or wants git/context/cost/rate-limit/telemetry info in the Claude Code status bar.

1 7d ago B 67 tokens original MIT