Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Import a workspace template. Supports preview (dry-run), collision handling (rename/skip/replace), and secret requirements tracking. Accepts templates from local files, URLs, or GitHub repositories. The counterpart to /export. Triggers on: "import", "load template", "install workspace"
.claude/skills/miosa-osa-import/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 831% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 866% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 231% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 228% | 0% |
> Import a workspace template with collision handling and secret tracking.
Stand up a workspace from an exported template. Handles the full import lifecycle: preview what will be created, detect collisions with existing files, resolve conflicts, prompt for required secrets, and provision the workspace. Supports templates from local files, URLs, or GitHub repositories.
bash# Import from local file /import templates/research-team.yaml # Import from URL /import --url "https://example.com/templates/research-team.yaml" # Import from GitHub /import --github "org/repo/templates/research-team.yaml" # Preview without importing /import templates/research-team.yaml --dry-run # Import with variable overrides /import templates/research-team.yaml --var project_name="MIOSA Analysis" # Import with collision strategy /import templates/research-team.yaml --on-collision rename # Import into specific directory /import templates/research-team.yaml --target /workspace/new-project
| Flag | Type | Default | Description | |------|------|---------|-------------| | <template> | positional | — | Local file path to template | | --url | string | — | URL to fetch template from | | --github | string | — | GitHub path: owner/repo/path (uses default branch) | | --dry-run | flag | false | Preview what would be created | | --var | key=value] | — | Variable overrides | | --on-collision | enum | ask | ask (interactive), rename (suffix), skip, replace | | --target | path | . | Target directory for workspace | | --skip-secrets | flag | false | Skip secret prompts (use placeholders) | | --skip-launch | flag | false | Import structure only, don't launch agents |
--var overrides. For unresolved variables without defaults, prompt user.--skip-secrets).--on-collision:ask: prompt for each collisionrename: add suffix to conflicting items (e.g., researcher-2)skip: keep existing, skip conflictingreplace: overwrite existing with template version--skip-launch, run /launch on the imported workspace.markdown## Import Preview: research-team-v2 **Source:** templates/research-team.yaml **Target:** /workspace/miosa-analysis/ ### Will Create | Item | Type | Detail | |------|------|--------| | SYSTEM.md | file | Workspace system prompt | | agents/researcher/ | directory + agent | Research Analyst (claude-sonnet, 80K budget) | | agents/analyst/ | directory + agent | Data Analyst (claude-sonnet, 60K budget) | | agents/writer/ | directory + agent | Content Writer (claude-haiku, 30K budget) | | inbox/ | directory | Empty inbox | | processed/ | directory | Empty output | | ops/ | directory tree | observations, metrics, budget | ### Variables | Variable | Value | Source | |----------|-------|--------| | project_name | "MIOSA Analysis" | --var override | | budget | 200000 | template default | ### Secrets Required | Secret | Status | |--------|--------| | ANTHROPIC_API_KEY | found in env | | OPENAI_API_KEY | MISSING — will prompt | ### Collisions | Item | Conflict | Resolution | |------|----------|------------| | agents/researcher/ | directory exists | will rename → researcher-2 | Proceed? (y/n)
Imported: research-team-v2 → /workspace/miosa-analysis/
Created: 14 items (3 agents, 5 dirs, 6 files)
Skipped: 0
Renamed: 1 (researcher → researcher-2)
Replaced: 0
Secrets: 1 still needs OPENAI_API_KEY (set env var to enable Codex adapter)
Status: launched (3 agents running)--url and --github sources)/launch — Workspace activation after import/export — Sister skill that produces importable templates| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,693 | 9,201 | -28% | 1 | 1 | 0% | 2,393 | 3,084 | +29% | 0 | 0 | — |
case-02 | fail→pass | 7,636 | 7,504 | -2% | 1 | 1 | 0% | 296 | 2,757 | +831% | 0 | 0 | — |
case-03 | fail→fail | 9,251 | 5,087 | -45% | 1 | 1 | 0% | 1,374 | 2,221 | +62% | 0 | 0 | — |
case-04 | fail→fail | 16,160 | 21,335 | +32% | 1 | 1 | 0% | 315 | 2,749 | +773% | 0 | 0 | — |
case-05 | fail→pass | 3,862 | 3,603 | -7% | 1 | 1 | 0% | 201 | 1,942 | +866% | 0 | 0 | — |
case-06 | fail→pass | 26,418 | 13,363 | -49% | 1 | 1 | 0% | 1,097 | 3,628 | +231% | 0 | 0 | — |
case-07 | fail→pass | 7,044 | 5,745 | -18% | 1 | 1 | 0% | 694 | 2,278 | +228% | 0 | 0 | — |
case-08 | fail→pass | 4,914 | 3,827 | -22% | 1 | 1 | 0% | 271 | 1,866 | +589% | 0 | 0 | — |
case-09 | fail→pass | 13,856 | 7,026 | -49% | 1 | 1 | 0% | 2,390 | 2,670 | +12% | 0 | 0 | — |
case-10 | fail→pass | 12,976 | 7,262 | -44% | 1 | 1 | 0% | 2,450 | 2,706 | +10% | 0 | 0 | — |
case-11 | fail→pass | 5,169 | 3,512 | -32% | 1 | 1 | 0% | 938 | 1,899 | +102% | 0 | 0 | — |
case-12 | fail→pass | 7,414 | 4,883 | -34% | 1 | 1 | 0% | 461 | 2,145 | +365% | 0 | 0 | — |
case-13 | fail→fail | 6,932 | 8,928 | +29% | 1 | 1 | 0% | 522 | 2,688 | +415% | 0 | 0 | — |
case-14 | fail→pass | 4,585 | 5,430 | +18% | 1 | 1 | 0% | 248 | 2,290 | +823% | 0 | 0 | — |
case-15 | pass→pass | 26,510 | 15,038 | -43% | 1 | 1 | 0% | 4,206 | 3,799 | -10% | 0 | 0 | — |
case-16 | fail→pass | 8,160 | 21,873 | +168% | 1 | 1 | 0% | 543 | 2,781 | +412% | 0 | 0 | — |
case-17 | pass→fail | 11,072 | 8,164 | -26% | 1 | 1 | 0% | 1,995 | 3,047 | +53% | 0 | 0 | — |
case-18 | fail→pass | 9,928 | 2,136 | -78% | 1 | 1 | 0% | 1,590 | 1,525 | -4% | 0 | 0 | — |
case-19 | fail→pass | 13,668 | 2,252 | -84% | 1 | 1 | 0% | 2,396 | 1,642 | -31% | 0 | 0 | — |
case-20 | pass→pass | 22,288 | 3,349 | -85% | 1 | 1 | 0% | 3,221 | 1,740 | -46% | 0 | 0 | — |
case-21 | pass→pass | 17,740 | 8,255 | -53% | 1 | 1 | 0% | 2,945 | 2,716 | -8% | 0 | 0 | — |
case-22 | fail→fail | 10,595 | 5,714 | -46% | 1 | 1 | 0% | 1,831 | 2,326 | +27% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 12 counted toward the lift figure. The other 10 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +56 percentage points is the difference between those two pass rates over the 12 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.