Huruikagi/pc-build-planner / .claude/skills/sb-configure
sb-configure
VerifiedConfigure or explicitly update SpecBind for a project and complete the resulting aftercare. Use for an explicit SpecBind binary or project-asset update request, initial post-install review, or later changes to Agents, role models, artifact language, project instructions, templates, shared Rules, operational adapters, Steering, or existing artifacts affected by configuration changes.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.5 KB |
| references/adapters.md | markdown | 3.1 KB |
| references/aftercare.md | markdown | 2.4 KB |
| references/installation-and-agents.md | markdown | 2.6 KB |
| references/migrations/descriptions.md | markdown | 2.2 KB |
| references/rules.md | markdown | 1.6 KB |
| references/steering.md | markdown | 1.2 KB |
| references/templates-and-reconciliation.md | markdown | 5.1 KB |
| references/update.md | markdown | 8.2 KB |
references/templates-and-reconciliation.md
View on GitHub# Templates and existing-artifact reconciliation
Use this procedure for Spec, Steering, or Roadmap templates, named creation
outputs, Design-template selection, and alignment of existing artifacts.
## Inspect and edit
List the applicable scope and read the selected raw template before editing:
```sh
specbind template list <scope>
specbind template read <scope> <selector>
```
For a Spec target that already exists, resolve its exact path:
```sh
specbind template resolve spec <spec> <selector>
```
Write only to the reported project-owned template path. Preserve the OKF
profile, selector, output path, machine-owned Front Matter boundary, and every
complete scoped-instruction block.
Every `{{name}}` body reference needs exactly one `create output=name`
instruction. Produce no output while editing the template. Each output may be a
short string or a Markdown fragment, and references are forbidden in Front
Matter. Do not restore a removed CLI renderer or implement a second template
language.
## Design template set
New Design and Steering templates require a non-empty single-line literal
Front Matter `description` describing their actual recurring independent
responsibility. Technology labels alone do not establish that responsibility.
Validate it with selector, output path, maintain/consume guidance, and (for
Design) `design-template-selection` in the same configuration transaction.
Existing major-one templates without descriptions remain readable. A template
edit does not add descriptions to live artifacts; use the separately confirmed
reconciliation procedure below. Requirements templates guide the author to
replace their generic description with the current Spec's responsibility.
Treat every discovered `design/<artifact_id>` template and
`design-template-selection` as one configuration transaction. Classify every
candidate exactly once as `required`, `conditional`, or `disabled`; give every
conditional entry an applicability condition. Re-run both the template list and
Rule read before completion.
Before adding a candidate, compare current Steering, repository facts, the main
Design, and the shared Design and Contract Rules. Prefer updating those existing
surfaces when the proposed content is a common heading, convention, or
cross-cutting judgment. Add a candidate only for a durable responsibility that
needs distinct recurring design decisions and traceability; a technology label
or framework presence alone is insufficient.
For example, user-visible Web or mobile behavior normally selects the existing
UI candidate. API compatibility or infrastructure conventions normally begin
in Steering and the Design or Contract Rules. A project may add `design/api` or
`design/infrastructure` when those responsibilities repeatedly need independent
design treatment. Their selection conditions must describe the responsibility,
not merely the presence of a technology.
If a request names a future technology but the current Steering and repository
facts do not establish an independent recurring responsibility, present the
existing-surface and conditional-candidate options and ask the maintainer to
choose. Do not infer that a request for possible future API or infrastructure
work authorizes a new candidate.
## Default effect
A template edit affects future materialization only. State explicitly that no
existing artifact changed, then offer reconciliation. Declining completes the
template change without touching live artifacts.
## Reconciliation preview
When accepted for preview:
1. Preserve the pre-change template from Git and compare it with the new one.
2. Enumerate Specs through `specbind spec list`; use `template resolve` for the
exact target of each applicable selector. Use the corresponding Steering or
Roadmap inventory for those scopes.
3. Read each live managed Markdown artifact through its owning CLI read surface.
4. Classify the proposal as `format-only`, `instruction-update`, `structural`,
`semantic`, or `conflict`.
5. Present a per-artifact diff or precise change summary, preserved identities,
and lifecycle effect without writing.
Template provenance is not stored in live artifacts. A matching selector and
path makes an artifact a candidate, not proof that it was materialized from the
previous template.
## Apply only a separately confirmed reconciliation
Preserve authored semantic content, Requirement and Contract identities,
artifact IDs, machine-owned fields, and durable instruction blocks unless the
preview explicitly covers an instruction update. Route semantic changes through
the ordinary Requirements, Design, Steering, or other owning Skill.
Never directly rewrite `spec.yaml`, `tasks.yaml`, Gate or completion evidence,
released archives or logs, or CLI-owned Roadmap Front Matter.
Artifact reconciliation and any resulting review, approval, or completion rerun
are separate confirmations. Byte changes can stale fingerprints even when the
agent classifies the edit as format-only.
## Verify
Re-list and re-read the template. For reconciled artifacts, run their parser,
traceability, status, and lifecycle checks. Fix structural invalidity; do not
invent semantic approval to make a check pass.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |