菜单
Dispatch the next ready porting candidate from the catalog. Session-resilient — dispatches ONE port per invocation. Re-run to continue the batch.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Capture a lesson learned from a porting session and route it to the right enforcement mechanism. Use when a bug, mistake, or process failure occurs that should be prevented in future sessions.
Publish a news entry to the amiport site. Appends to site/data/news.json, validates, deploys via site-manager, and clears the activity cache. Use when announcing a release, project update, milestone, or behind-the-scenes note.
Analyze C source code for Amiga portability issues. Scans headers, system calls, and language features to produce a structured porting report. Use when starting a new port or evaluating whether a project can be ported.
Transform C source code for Amiga compatibility. Replaces POSIX calls with Amiga equivalents or posix-shim wrappers. Use after analyze-source has identified issues.
Dispatch multiple ports from the catalog in parallel using specialized agents. Runs the same pipeline stages as /port-project but overlaps them across ports for ~3x throughput.
Port an entire C project to AmigaOS. Runs the full pipeline — analyze, transform, build, test, package. Use to port a new project from scratch.
| name | batch-port |
| description | Dispatch the next ready porting candidate from the catalog. Session-resilient — dispatches ONE port per invocation. Re-run to continue the batch. |
| user_invocable | true |
Dispatch the next best porting candidate from the catalog.
/batch-port # Next best Cat 1 candidate for a1200_accel
/batch-port cli a1200_accel # Explicit category + profile
/batch-port cli stock_a1200 # Different hardware profile
/batch-port scripting a1200_accel # Category 2 candidates
This skill dispatches ONE port per invocation. After the port completes (or the session ends), re-run /batch-port to dispatch the next candidate. The catalog.json status field IS the queue — it persists across sessions.
python3 scripts/catalog-score.py --score to re-score all candidates (accounts for any shim changes since last run)python3 scripts/catalog-score.py --next 1 --category <cat> --profile <prof> to get the best candidatepublishable field — skip if falsestatus: "in_progress" in catalog.json (atomic write)catalog-engineer agent with the candidate details:
/port-project for the candidate--unattended mode (auto-answer decisions)ported[] in catalog.jsonstatus: "failed" with error reasoncli (Cat 1 — the only fully autonomous category)a1200_accel (the standard FS-UAE test config)If a candidate has status: "in_progress" (stale from a crashed session), treat it as a retry candidate. Re-dispatch it before selecting a new one.
Run python3 scripts/catalog-score.py --status to see the updated dashboard.
Run python3 scripts/catalog-score.py --diff data/catalog-before.json to see what changed (if you saved a snapshot before the batch).