Menu
Bootstrap architectural memory for a repo — generates starter cards from a structured brief
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | wipnote:arch-bootstrap |
| description | Bootstrap architectural memory for a repo — generates starter cards from a structured brief |
Use this skill to populate a project's architectural memory from scratch. It runs the bootstrap command, reads the brief, and authors starter arch cards via the validated add path.
Trigger keywords: bootstrap arch, bootstrap architectural memory, starter cards, arch bootstrap, initialize arch cards
wipnote arch bootstrap
Read the full output. It contains:
For each major subsystem, invariant, or hazard you identify from the brief, run:
wipnote arch add <slug> \
--kind <kind> \
--created-by agent \
--paths "<glob1>,<glob2>" \
--body "<body up to 120 words>"
Card kinds:
subsystem-map — what a subsystem does and its boundariesinvariant — a constraint that must always holdhazard — a known failure mode or sharp edgedecision — a recorded architectural decisionRules:
** for recursive match (e.g. internal/**, cmd/wipnote/*.go)Aim for 3–8 cards covering the top subsystems visible in the hotspot list.
Do NOT mechanically import CLAUDE.md content — those are instructions, not architecture. Focus on structural facts: what systems exist, what invariants must hold, what hazards to avoid.
wipnote arch validate
Fix any reported errors (body too long, missing required fields, invalid kind).
wipnote arch list
wipnote arch show <slug>
Verify the cards describe real structural facts about the codebase, not process instructions.
wipnote orchestration patterns for AI-assisted development. Use when working on code in an wipnote project — provides delegation patterns, model selection, quality gates, and work tracking guidance. Activate when planning work, delegating to agents, debugging, building features, or managing tasks.
wipnote orchestration patterns for AI-assisted development. Use when working on code in an wipnote project — provides delegation patterns, model selection, quality gates, and work tracking guidance. Activate when planning work, delegating to agents, debugging, building features, or managing tasks.
wipnote orchestration patterns for AI-assisted development. Use when working on code in an wipnote project — provides delegation patterns, model selection, quality gates, and work tracking guidance. Activate when planning work, delegating to agents, debugging, building features, or managing tasks.
Plan development work using a triage-gated, blocks-first interview. Classify scope as trivial/standard/complex, then run 0/3/4 staged interview rounds through the current harness native ask-user tool when available, or plain conversation when it is not. Each stage elicits the VISUAL block first (file-tree, api-endpoint, data-model, wireframe/diagram) and derives the prose slice fields (what/why/done_when) from it — blocks are authored inline as the YAML is built, not in a post-pass. Produces slice-card YAML with grounded visual blocks and visible research provenance for external claims; pauses for human review; promotes approved slices to features. Use when asked to plan, create a development plan, or build a feature with design clarity first.
SECONDARY enrichment path — add or revise structured visual blocks (data-model, api-endpoint, file-tree, wireframe, diagram, tabs) on an EXISTING wipnote plan. Use for legacy plans drafted before blocks-first existed, or for ad-hoc later block additions/revisions — NOT for primary planning. New plans get blocks authored inline during the wipnote:plan interview (blocks-first); only reach for this skill when a plan already exists and needs blocks added or changed after the fact. Reads the live block catalog from wipnote plan blocks, adds grounded blocks to slices, regenerates the plan HTML, and links to the dashboard.
Generate a grounded visual recap for a work item, git range, or session. Runs wipnote recap to produce a self-contained HTML artifact grounded in the real diff, reads the artifact, writes an outcome narrative (secrets redacted), and links to the dashboard recap URL. Use when asked to recap, summarize, or visualize completed work.