Self-referential loop until task completion with configurable verification reviewer
Use when reviewing code (or your own plan) to allocate attention based on the danger of the change.
Use when deciding the lowest-cost context path for a mixed corpus, especially when choosing among direct reading, helper conversion, graph build, graph update, or graph query.
Use when a bug was recently introduced but you don't know which commit caused it.
Relay questions and tasks to a local Codex CLI using one-shot subprocesses per ask.
Vendor-neutral routing guide for choosing the right model tier by task type. Mechanical work uses a smaller/faster model; implementation uses a standard model; architecture,…
Use after fixing a bug to generate a blameless post-mortem summary for human review.
Use to isolate a bug from a large application into a standalone, runnable script or single test case.
Use when a user needs to build or refresh persistent graph memory from a mixed corpus and the right path may include graphify, incremental update, or helper conversion before…
Use when modifying dbt metrics or semantic models to ensure mathematical correctness and backwards compatibility.
Verify that all hyperparameters, metrics, and data references are properly logged.
Audit agent configuration files for security vulnerabilities and misconfigurations. Covers settings.json, .mcp.json, .codex/config.toml, AGENTS.md, hooks, plugin manifests, and…
Parallel execution engine for high-throughput task completion
Use to restructure code while guaranteeing that all existing tests continue to pass.
Use when designing or using MCP-backed structured code search with search, AST query, symbol inventory, and bounded extraction workflows.
Use when migrating APIs, libraries, or patterns across a large codebase. Ensures safe, step-by-step progress rather than risky mega-commits.
Deterministic 3-cycle loop for gathering codebase context before acting. Broad search → exact source and tests → target-specific docs and setup.
Use to identify and safely delete unused functions, classes, exports, and files.
Use when confronted with an unknown failure in CI or production to rapidly categorize the issue before deep debugging.
Use when evaluating prompts, LLM outputs, red-team suites, or model behavior with local eval configs and safe provider/cost controls.
Use when querying, ingesting, or maintaining a local RAG MCP corpus for semantic document retrieval with privacy controls.
Analyze BigQuery usage, identify cost hotspots, repeated failures, and practical optimization opportunities.
Use when designing or reviewing filesystem MCP access, path boundaries, allowed roots, method allowlists, and safe local file operations.
Evaluate metric and semantic model changes for BI/reporting breakage and business meaning drift.
Design agent tool sets with stable names, narrow schemas, deterministic output shapes, and explicit error paths. No catch-all tools unless unavoidable.
Audit documentation for broken file paths, outdated commands, and renamed variables.
Use before tagging a release or deploying to production to ensure all quality gates have passed.
Review SQL for business logic correctness, semantic drift, aggregation risk, and silent definition changes.
Check existing repo capability, external libraries, MCP options, and maintenance risk before writing custom code. Decide adopt/wrap/build with explicit criteria.
Use to audit and remove unused or redundant third-party dependencies from package manifests.
Use to enforce consistent naming conventions and file structures across a project without changing business logic.
Use when designing, running, debugging, or hardening deterministic eval suites for agent skills, prompts, tool workflows, or MCP-backed cases.
Use continuously during long tasks. Teaches how to read less, output less, and keep the LLM context window lean and fast.
Use when completing a task or running out of context limit. Ensures the next session or human engineer has exactly what they need to resume work instantly.
Audit whether a dbt incremental model uses the right incremental strategy for the repo, the data shape, and the operational constraints.
Use at the beginning of a new task. Ensures you fully understand the requirements, boundaries, and acceptance criteria before writing code.
Use when writing technical documentation that needs to be readable by both humans and AI models, converting existing docs to HADS format, validating a HADS document, or optimizing…
Use when mining PR review comments and text diffs for reusable writing, documentation, tone, and editorial improvement patterns.
Analyzes ML training scripts to enforce seed setting, deterministic operations, and environment tracking for exact reproducibility.
Use to collapse over-engineered abstractions, remove unnecessary layers, or consolidate redundant logic.
Use when the task is to understand an unfamiliar codebase, locate key entry points, or summarize architecture before editing.
Use when optimizing agent runtime loops, card packs, MCP session lifecycle, tool-call count, or multi-agent orchestration patterns.
Compact context at logical phase boundaries — after research, after planning, after debugging — rather than mid-task. Preserves useful state while clearing noise.
Maintain short, focused Markdown files per subsystem to provide agents with isolated context.
Ensure the project README provides immediate, exact commands for setup, testing, and deployment to help agents and humans bootstrap quickly.
Full autonomous execution from idea to working code
Standardize the reporting of model metrics to ensure statistical rigor and business relevance.
Use when validating rendered web pages, local dev servers, browser automation, screenshots, forms, auth sessions, or UI evidence with strict browser safety boundaries.
Relay questions and tasks to a persistent local Gemini ACP session with cross-turn context.
Write state-restoration documents for passing tasks between agents or engineers.
Diagnose NaN losses, out-of-memory errors, and shape mismatches in deep learning or ML pipelines.
Inspect changed dbt assets, estimate blast radius, identify missing tests, and recommend the narrowest safe validation plan.
Use when creating or reviewing red-team eval plugins, attack templates, grader rubrics, safety fixtures, or model-risk test metadata.
Use when diagnosing agent session history, interrupted tool loops, missing tool results, timing bottlenecks, or subagent trace correlation.
Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy.
Use when starting work in a repository with Agent Powerups installed, when a task may match a reusable local skill, command, workflow, hook recipe, AGENTS.md template, or MCP…
Use before submitting a PR or considering a task done to evaluate the 'blast radius' of your changes.
Use when creating or modifying dimensional dbt models in warehouse-backed analytics projects. Covers a four-layer warehouse architecture (sources/staging/core/marts), naming…
Use when a CI pipeline fails to extract the actual error from thousands of lines of logs.
Use when debugging complex runtime failures, distributed systems, or issues where a local debugger cannot be attached.