Huruikagi/pc-build-planner / .agents/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 | 6.4 KB |
| agents/openai.yaml | config | 206 B |
| 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/installation-and-agents.md
View on GitHub# Installation and Agent configuration
Use this procedure for selected Agents, artifact language, root project
instructions, or role capability overrides.
## Inspect
Read `.specbind.json` only after `specbind configuration show` has validated it.
Treat these as the supported configuration fields:
- `specDir`: fixed after initial installation in v1;
- `language`: `en` or `ja`;
- `agents`: `claude-code`, `codex`, or `generic`;
- optional `projectInstructions: true`;
- optional `adoption: true` while the temporary `sb-adopt` capability is installed; and
- optional `agentRoles` for Codex or Claude Code.
`generic` has no product role definitions and cannot own `agentRoles`.
## Add or refresh
Agent selection through install is additive. Preview the exact current plan,
then apply the same inputs:
```sh
specbind install --dry-run [--agent <agent>] [--language <language>] [--project-instructions] [--with-adoption|--without-adoption]
specbind install [--agent <agent>] [--language <language>] [--project-instructions] [--with-adoption|--without-adoption]
```
Do not omit a persisted true `projectInstructions` choice when constructing an
explicit invocation. Omit the adoption flags to preserve the persisted choice;
use `--with-adoption` only for explicit initial reverse establishment and
`--without-adoption` only for its guarded retirement or retry. Never edit
installed Skill files, generated role files, or managed root-instruction blocks
directly.
For `agentRoles`, edit only `.specbind.json`, preserve unrelated fields, run the
dry run, and reinstall. A configured model that the host cannot start is a
configuration or environment failure; do not silently substitute another.
## Remove
One-Agent removal is plan first and keeps shared surfaces required by remaining
Agents:
```sh
specbind remove-agent <agent>
specbind remove-agent <agent> --apply
```
Present the exact plan and obtain confirmation before `--apply`. Removing the
last Agent routes to guarded project uninstall and its explicit durable-
knowledge choice; configuration approval alone never authorizes uninstall.
## Language
Changing the configured language refreshes product-managed renderings but never
overwrites existing project-owned templates, Rules, adapters, Steering, or live
artifacts. Enumerate those retained files as aftercare. Offer translation as a
separate previewed content change; never describe a mixed-language project as
automatically migrated.
## Verify
Re-run:
```sh
specbind configuration show
specbind install --dry-run
```
The final dry run must contain no unexpected create or replace action. Report
host model availability separately when it cannot be proved mechanically.
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: 322k | 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: 322k | 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: 322k | Updated | Claude CodeCodex | Verified | Install |