Agent Claude Code
Drafts Architecture Decision Records (ADRs) from a decision prompt or discussion. Structures rationale, alternatives considered, consequences, and reversibility in a format suitable for committing to a decisions/ directory.
Agent Claude Code
AI Engineering review of code touching LLM interactions, prompt construction, context management, agent architecture, and AI-specific security. Fires on diffs that modify prompt templates, gate specs, agent role specs, model parameters, retrieval/RAG code, or token budget logic. Strict mode - REVISE blocks commits.…
Agent Claude Code
Generates a concise current-state architecture overview with Mermaid diagrams, entry points, key abstractions, data flow, and drift observations. Catches drift between the mental model and actual code. Use on-demand or at project milestones -- NOT a code review, this is a map. Context: Just merged a large refactor.…
Agent Claude Code
Creative research and solution design agent. Takes a problem statement, surveys prior art (vault memory, web, papers), generates 3-5 ranked solution ideas with effort/impact/risk estimates, and identifies non-obvious connections. Use when stuck on a challenge, exploring design alternatives, or wanting creative input…
Agent Claude Code
Analyzes a CV or career history against a target role to identify skill gaps, narrative gaps, and positioning gaps -- with specific, prioritized actions to close each gap before applying.
Agent Claude Code
Fast read-only code exploration with codegraph intelligence. Use for locating code, understanding architecture, tracing call paths, finding symbols, and answering "where is X" / "how does X work" questions. Replaces built-in Explore with codegraph-first tool selection.
Agent Claude Code
Adversarial review of a contract diff -- identifies clauses that shift risk, limit rights, or create obligations not visible in isolation. Surfaces what changed and what it means, not just what text was altered.
Agent Claude Code
Reviews all user-facing text — error messages, help text, status indicators, onboarding copy, system messages. Ensures text is clear, human, actionable, and consistent in tone. NOT about code quality — about how the product speaks to the user. Advisory (not a commit gate). Use after changes that add or modify…
Agent Claude Code
Pre-mortem analysis of a planned decision or project -- imagines failure, traces failure modes to root causes, and produces a risk-ranked list of preventable failure paths with specific mitigations.
Agent Claude Code
Weekly background agent that scans docs/ for stale content and the codebase for pattern drift, then opens targeted fix-up PRs. Runs automatically via the scheduler. One concern per PR. Max 3 PRs per run.
Agent Claude Code
Audits an evaluation setup (benchmark, A/B test, or model comparison) for methodology errors that would invalidate results -- sampling mismatches, scope leakage, judge reliability, and baseline validity.
Agent Claude Code
Decomposes a technical concept into layered explanations using the Feynman technique -- analogy first, mechanics second, where the analogy breaks third, edge-cases fourth, connection to existing knowledge fifth.
Agent Claude Code
General-purpose agent for researching complex questions, searching for code, and executing multi-step tasks. Use when no specialized agent fits. Has full tool access including codegraph for code intelligence.
Agent Claude Code
Formats Hebrew academic content for XeLaTeX compilation with correct RTL, polyglossia, and font configuration. Validates OOXML bidi rules for .docx output. Produces compilable LaTeX or corrected OOXML fragments.
Agent Claude Code
Structured end-to-end trace to find the FIRST broken link in a specific claim's dependency chain. Single-claim depth probe — NOT a breadth reviewer. Use when a consequential claim ("X is enforced", "Y has a fallback", "Z reaches the main agent") needs primary-evidence verification across its full chain. Advisory…
Agent Claude Code
Parallel literature and evidence scout that retrieves, classifies, and synthesizes sources using a structured evidence-quality taxonomy. Prioritizes empirical data over opinion; surfaces contradictions explicitly.
Agent Claude Code
Authors the discriminating red-green test (the "oracle") for a high-blast-radius change FROM THE SPEC, blind to the implementation, BEFORE any code is written. Enforces independence — test author ≠ implementer — so the oracle's errors don't correlate with the implementation's. Advisory role (not a commit gate)…
Agent Claude Code
Independent second opinion on a development plan BEFORE implementation. Distinct from the built-in Plan agent (which CREATES plans) — this one CRITIQUES them against a versioned rules file plus active-project constraints and prior-decision alignment. Use when the user (or you) just drafted a plan for a non-trivial…
Agent Claude Code
Software architect agent for designing implementation plans. Use when you need to plan implementation strategy for a task. Returns step-by-step plans, identifies critical files, and considers architectural trade-offs. Read-only — no file modifications.
Agent Claude Code
Post-implementation test strategy reviewer. Evaluates whether changes have adequate test coverage, identifies untested edge cases, and generates concrete test scenarios. NOT a code reviewer — focuses on what ISN'T tested. Advisory (not a commit gate). Use after implementation when you want to verify test completeness…
Agent Claude Code
Structured adversarial review of a claim, result, or output -- stress-tests assumptions, checks for confounds, flags overreach, and rates confidence. Complements (not replaces) domain review.
Agent Claude Code
Cross-session pattern analyzer. Reads the last N session logs and produces a dated retrospective artifact with root cause hypotheses, deferred task tracking, behavioral drift checks, and verifiable recommendations with confidence levels and testable predictions. Each recommendation gets a unique ID tracked across…
Agent Claude Code
Architecture-level threat review before implementing any feature touching auth, data storage, external APIs, or trust boundaries. Reviews design against STRIDE/OWASP -- NOT code-level security (code-reviewer covers that). Use when the plan involves credentials, sessions, inter-service calls, sensitive data flows, or…
Agent Claude Code
Post-implementation UX audit of user-facing changes. Evaluates interaction patterns against usability heuristics, competitive benchmarks, and edge-case scenarios. Produces a prioritized punch list of experience issues — not code quality, but whether it feels right to a user. Advisory (not a commit gate). Use after…
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: