AI Engineering

Spec Kitty routes standalone requests through a governance context, opens an Op record, and closes it with the real outcome after the work is done. Use it when a user wants Spec Kitty involved outside a full mission workflow.

Spec Kitty Agent Config configures agent profiles, host-specific paths, command installation, and model/tool routing for Spec Kitty. Use it when setting up Codex, Claude, OpenCode, or another agent to work with Spec Kitty.

Spec Kitty Bulk Edit Classification recognizes when a mission is a bulk edit and drives the occurrence-classification guardrail so users never have to know it exists. Use it during specify or plan for renames, terminology migrations, and path/module moves.

Spec Kitty Charter Doctrine runs charter interview, generation, context, and sync workflows for project governance in Spec Kitty 3.x. Use it to access doctrine artifacts programmatically, resolve agent profiles, and load action-scoped governance context iteratively.

Spec Kitty Glossary Context curates canonical terminology across Spec Kitty missions, detects term drift, and enforces consistent language in mission artifacts. Use it to update the glossary, resolve ambiguity, or review terminology consistency before generation.

Spec Kitty Implement Review orchestrates the implement-review loop for Spec Kitty work packages using any configured coding agent. Use it to dispatch implementation and review tasks, handle rejection cycles, and sequence work packages by dependency.

Spec Kitty Mission System explains the four built-in mission types in Spec Kitty (software-dev, research, plan, documentation), their step contracts, template resolution, and doctrine artifacts. Use it to choose a mission or understand how missions structure work.

Spec Kitty REASONS Canvas drives authoring and review of the REASONS canvas for missions using Structured-Prompt-Driven Development (SPDD). It detects charter activation, loads artifacts, generates or updates the seven-section canvas, and runs drift review.

Spec Kitty Runtime Next manages the runtime phase of spec-driven development, providing a taxonomy for runtime results and recovering from blocked states. Use it when an agent needs to interpret runtime outcomes and continue work after interruptions.

Spec Kitty Runtime Next drives the canonical spec-kitty next control loop so an agent can advance missions through step decisions, WP iteration management, and doctrine-aware context loading. Use it when you need to determine the next action, unblock a stalled mission, or interpret runtime outcomes.

Spec Kitty SPDD Reasons drives authoring and review of REASONS Canvases for Spec Kitty missions that opted into Structured-Prompt-Driven Development. Use it to generate or update the canvas, compile per-WP summaries, or compare implementation against the canvas in review mode.

Planning creates structured plan documents and tracker items using YAML frontmatter for bugs, tasks, ideas, and features. Use it to plan a feature, track progress, log work items, or organize project work.