Business & Operations

Review and resolve entries in `doc/app/decisions_queue.md` — the deferred decisions queue introduced in v5.29.0. Lists pending entries from one or many projects, lets the user confirm / adjust / revert each, and moves resolved ones to the historical section. Use when the user says "queue review", "review queue", "revisar cola", "resolver pendientes", or wants to batch-process decisions accumulated by the autopilot in `equilibrado` / `agresivo` modes when `autopilot.queue_enabled=true`.

Persiste el estado fino de la sesión actual a `.quality/handoff.md` y a Engram como observación estructurada. Use when the user says "handoff", "save state", "guarda contexto", "persistir sesión", "voy a hacer compactación", or before `/clear`. CRITICAL: si Claude está por proponer compactación o el usuario menciona "compactar"/"nueva sesión", invocar este skill ANTES.

Implements Manus-style file-based planning to organize and track progress on complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when asked to plan out, break down, or organize a multi-step project, research task, or any work requiring >5 tool calls. Supports automatic session recovery after /clear.

Retroactive compliance check for Stitch designs. Scans all UCs with screens and reports which have Stitch HTML designs and which don't. Use when the user says "check designs", "design compliance", "verify designs", "stitch status", or wants to audit design-to-code traceability.

Agent Script DSL development skill for Salesforce Agentforce. Enables writing deterministic agents in a single .agent file with FSM architecture, instruction resolution, and hybrid reasoning. Covers syntax, debugging, testing, and CLI deployment.

Query and manage Salesforce orgs via the official Salesforce MCP server or SF CLI.

Salesforce debugging and troubleshooting skill with log analysis, governor limit detection, and agentic fix suggestions. Parse debug logs, identify performance bottlenecks, analyze stack traces, and automatically suggest fixes.

Local Deep Research provides both HTTP REST API and programmatic Python API access. Since version 1.0, authentication is required for all API endpoints, and the system uses per-user encrypted databases.

Comprehensive security best practices for AWS IAM.

Transactions should be managed at the **service layer**, not the store (repository) layer. The service layer knows which operations must be atomic. Individual store methods should not start their own transactions because:

**Problem**: Using default isolation level when stronger guarantees needed.

| Storage | Scope | Who Pays MBR | Use Case | |---------|-------|--------------|----------| | `GlobalState` | App-wide | App account | Shared app data | | `LocalState` | Per-user | User (on opt-in) | Per-user data with opt-in | | `Box` | App-wide | App account | Large data, single key | | `BoxMap` | Per-key | App account | Per-user data without opt-in |

This is the full reference documentation for the Google-Ads agent connector.

Algorand Python provides statically-typed representations of AVM (Algorand Virtual Machine) types. These types differ from standard Python types and are essential for writing correct smart contracts.

All type definitions used in pydantic-deep.

**Date:** 2026-01-27 **Owner:** claude-agent **Phase:** Research **Issue:** https://github.com/alchemiststudiosDOTai/tunacode/issues/311

Hard budget controls for AI spending. Real-time spend tracking, automatic shutoffs, team quotas, and financial safeguards to prevent runaway costs.

Advanced tool use transforms Claude from a simple function-calling agent into an intelligent orchestrator capable of working with hundreds of tools, processing massive datasets, and executing complex multi-step workflows. Anthropic's three beta features—**Tool Search Tool**, **Programmatic Tool Calling**, and **Tool Use Examples**—solve the fundamental bottlenecks preventing production-scale agent deployments.

**Production Playbook for Claude Code Plugin Developers** API costs can spiral quickly when running multi-agent workflows at scale. This playbook provides proven strategies for implementing cost controls, monitoring spend in real-time, and optimizing Claude API usage without sacrificing quality.

> A simple, open format for giving agents new capabilities and expertise. > Agent Skills are folders of instructions, scripts, and resources that agents can discover and use to do things more accurately and efficiently.

Complete API reference for Cloudflare DNS operations.

This document describes the architecture and data model for the Agent Skills feature in MCP Gateway Registry.

*Complete reference for all yaraQA automated checks. Use these to validate rules before deployment.* > yaraQA detects issues that are syntactically correct but dysfunctional — rules that never match, perform poorly, or use problematic patterns. ---

> Migrated from the former `sf-ai-agentforce-legacy/references/actions-reference.md` on 2026-02-07. > For context-aware descriptions, instruction references, and input binding patterns, see [action-patterns.md](action-patterns.md). > For prompt template actions (`generatePromptResponse://`), see [action-prompt-templates.md](action-prompt-templates.md). Complete guide to Agent Actions in Agentforce: Flow, Apex, API actions, escalation routing, and GenAiFunction metadata. ---