Desenvolvimento

Restrict file edits to one directory for the session while debugging. CARL TRIGGERS: freeze edits, lock to this directory, freeze the codebase, edit lock. SOURCE: garrytan/gstack/freeze, integrated as gs-freeze on 2026-05-29.

GitHub の作業状況を収集して朝会・夕会を行う。 Use when the user says "standup", "morning standup", "evening standup", "朝会", "夕会", "振り返り", "daily standup", or invokes /standup. Supports morning (plan the day) and evening (reflect on today) modes. Supports multiple repository paths as arguments. Supports --export html option to export the report as an HTML file. Supports --template option to customize the report forma

Use when making any code change. Use when committing, branching, resolving conflicts, or when you need to organize work across multiple parallel streams.

Updated skill

Guide for using pytest-mock plugin to write tests with mocking. Use when writing pytest tests that need mocking, patching, spying, or stubbing. Covers mocker fixture usage, patch methods, spy/stub patterns, and assertion helpers.

Adversarial pressure-test of a decision, plan, or piece of work. Articulates the case FOR the current approach, then spawns a subagent to steel-man the best alternative, returns a synthesis with mandatory verdict. Invoke ONLY via /adversary — do NOT trigger on 'what do you think' or 'any concerns'.

Searches for non-coding RNA homologs and classifies RNA families using Infernal covariance model searches against the Rfam database. Identifies structured RNAs by sequence and secondary structure conservation. Use when querying sequences against Rfam, building custom covariance models for novel RNA families, or classifying non-coding transcripts by family.

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

Model interpretability and explainability using SHAP (SHapley Additive exPlanations). Use this skill when explaining machine learning model predictions, computing feature importance, generating SHAP plots (waterfall, beeswarm, bar, scatter, force, heatmap), debugging models, analyzing model bias or fairness, comparing models, or implementing explainable AI. Works with tree-based models (XGBoost, L

Debugs the Buttercup CRS (Cyber Reasoning System) running on Kubernetes. Use when diagnosing pod crashes, restart loops, Redis failures, resource pressure, disk saturation, DinD issues, or any service misbehavior in the crs namespace. Covers triage, log analysis, queue inspection, and common failure patterns for: redis, fuzzer-bot, coverage-bot, seed-gen, patcher, build-bot, scheduler, task-server

Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 → 4.6, 4.6 → 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a C

Side-route skill for compacting the current session into a transient handoff doc when no pipeline compression artifact fits — mid-skill, exploratory, side-route, or non-pipeline work. Not for routine inter-skill handoffs; use the calling skill's closing summary + **Next session:** line instead.