ai-build

Solid

Canonical implementation gateway: reads approved plan.md, resolves stack from manifest, deterministic-routes each task to its adapter, dispatches the build agent in an isolated worktree, runs TDD self-validation per task, then a single final quality loop with one bounded quality-remediation pass on the full changeset before /ai-pr. Trigger for 'go', 'start building', 'execute the plan', 'implement it', 'lets do this', 'build the plan', 'resume', 'continue'. Not without an approved plan; run /ai-plan first. Not for multi-concern specs needing decomposition; use /ai-autopilot instead. Not for a single function or subcomponent; use /ai-code.

AI & Automation 52 stars 3 forks Updated 4 days ago MIT

Install

View on GitHub

Quality Score: 89/100

Stars 20%

Recency 20%

100

Frontmatter 20%

Documentation 15%

100

Issue Health 10%

License 10%

100

Description 5%

100

Skill Content

# Build ## Purpose Execution engine for approved plans. Reads plan.md and tasks.md, dispatches one subagent per task (fresh context); each task self-validates via TDD; a single final quality loop runs verify+review plus one bounded quality-remediation pass on the full changeset before /ai-pr. If stuck: STOP and re-plan. ## When to Use - After `/ai-plan` produces an approved plan - To resume execution: `/ai-build --resume` - Never without an approved plan (run `/ai-plan` first) ## Process 0. **Preflight dependencies** -- verify `.ai-engineering/specs/plan.md`, `.agents/skills/_shared/execution-kernel.md`, `.agents/skills/ai-build/handlers/quality.md`, and `.agents/skills/ai-build/handlers/deliver.md` exist. If any are missing: STOP and report the exact missing path(s). Never improvise missing orchestration logic. 0b. **Executor route gate (spec-145)** -- read plan frontmatter `status` and `execution_route.executor`. If `status` is not `approved`, STOP for operator approval. If `executor: autopilot`, refuse this `/ai-build` run, print `execution_route.safe_next_command` (normally `/ai-autopilot`), and STOP. If route metadata is absent, continue with legacy behavior. 1. **Board sync (in_progress)** -- read `.ai-engineering/specs/spec.md` frontmatter `refs`; for each work item ref where the hierarchy rule is not `never_close` (i.e., user_stories, tasks, bugs, issues), invoke `/ai-board sync in_progress <work-item-ref>`. Fail-open: do not block DAG construction if this fails...

Details

Author: arcasilesgroup
Repository: arcasilesgroup/ai-engineering
Created: 4 months ago
Last Updated: 4 days ago
Language: Python
License: MIT

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Solid

ai-plan

Decomposes an approved spec into a phased execution plan with bite-sized tasks, agent assignments, and gate criteria — the contract /ai-build executes. Trigger for 'break this down', 'create a plan', 'what tasks do we need', 'lets start implementing', 'scope changed re-plan'. Hard gate: user approves before /ai-build can run. Not for ambiguous requirements; use /ai-brainstorm instead. Not for execution; use /ai-build instead.

52 Updated 4 days ago

arcasilesgroup

AI & Automation Listed

build

Plan then implement with an explicit approval gate between phases — invoked via /build

0 Updated 2 weeks ago

puukis

AI & Automation Solid

ai-autopilot

Delivers large multi-concern specs and backlog runs autonomously: decomposes specs into sub-specs (or normalizes work items into a backlog DAG), deep-plans with parallel agents, builds a dependency DAG, implements in waves, runs a single final quality loop with one bounded quality-remediation pass (verify+guard+review on full changeset), delivers via PR. Trigger for 'implement spec-NNN end to end', 'autopilot this', 'autonomous delivery', 'decompose and ship', 'run the backlog', 'execute these GitHub issues', 'process the sprint backlog'. Invocation is the approval gate. Not for small or single-concern tasks; use /ai-build instead. Not for ambiguous requirements; use /ai-brainstorm first.

52 Updated 4 days ago

arcasilesgroup