メニュー
Author one runtime-neutral catalog asset that is adapter-translated per runtime (OpenCode/Claude) — single source, no duplicate per-runtime files.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | runtime-author |
| description | Author one runtime-neutral catalog asset that is adapter-translated per runtime (OpenCode/Claude) — single source, no duplicate per-runtime files. |
Every runtime asset (agent body, skill, workflow, command, template, hook) is authored once in the catalog as runtime-neutral markdown. The OpenCode adapter and the Claude Code adapter each translate it for their runtime at export time. This skill ensures the source is written so both adapters produce valid output — preserving single source (R-006) and the byte-parity moat.
When creating or modifying any catalog asset that will be rendered to a live runtime: agents, skills, templates, commands, workflows, hooks.
aspis validate-runtime (agents) and resolve
every broken reference before continuing.aspis export --dry-run; confirm no render
errors for OpenCode or Claude Code. Fix the catalog — never patch the
live file.aspis export --check; the catalog must
regenerate the live runtime byte-for-byte. DRIFT means a hand-edit or a
non-deterministic render — refuse the change until CLEAN.permission: block survives translation (known past asymmetry — a single
stripped field is the worst kind of small change).validate-runtime pass and an --dry-run render with no errors.tools: vs permission:) —
breaks the other adapter's render.--dry-run or --check; trusting one adapter without testing both..opencode/, .claude/) directly instead
of the catalog source — guarantees the next parity check fails.Keep commits atomic — split a mixed change set into scoped, reviewable commits.
The one-time procedure to take an exported project to live — understand it, confirm the name/goal/stack with the user, run the deterministic `aspis bootstrap`, enrich the judgment files (AGENTS.md, ARCHITECTURE, context), verify, and let the package self-clean. Followed by the bootstrap agent.
Keep per-subsystem architectural intent current through the planning loop — read before designing, record an impact report on change, confirm with the user, apply a dated update, and verify the build against approved intent.
Audit a plan's task dependency graph for structural integrity — circular dependencies, missing prerequisites, orphan tasks, and dependency classification (hard/soft/warning). Produces a pass/warn/fail audit report per dependency so planners catch graph errors before build starts.
Every editing agent should start on a clean working tree so parallel work never collides.
Verify hooks ran, no secrets, protected paths untouched, and commit message valid before committing. Owned by the reviewer.