菜单
Scan open pull requests and reconcile the needs-review label. Use before running agentic-pr-static-review to find PRs needing a review pass. Outputs JSON with reviewed, needs_review, labelled, clean, incomplete, and error arrays.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/warpfront/hipfire --skill agentic-pr-discovery命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Port hipfire compute kernels to a new RDNA / CDNA architecture (gfx1201/gfx1200/gfx94x/gfx1150/etc.). Use when adding support for a new GPU arch, fixing arch-specific kernel codegen failures (e.g. "Cannot select intrinsic %llvm.amdgcn.wmma..."), or refactoring rdna-compute arch-conditional dispatch. Captures the workflow for WMMA/MFMA ports, ArchCaps routing, channel/speed validation selection, contributor onboarding, and known correctness traps. Triggers on phrases like "port to gfx12", "9070 XT support", "R9700 support", "WMMA gfx12", "Cannot select intrinsic wmma", "amdgcn.wmma", "new arch port", "cross-arch kernel".
Triage and repair hipfire runtime failures such as daemon hangs, stale serve.pid, port 11435 conflicts, ROCm include-path problems, missing precompiled kernels, VRAM OOM, kernel JIT failures, and multi-turn recall regressions. Use after diagnostics identify a likely runtime issue or when the user asks to fix a broken hipfire serve/run flow.
Run and interpret hipfire GPU diagnostics for ROCm/HIP bring-up, missing kernels, test_kernels failures, inference smoke failures, and install/runtime environment problems. Use when a user asks to diagnose hipfire, check GPU readiness, run baseline tests, or explain diagnostic output.
基于 SOC 职业分类
正在显示 SKILL.md
| name | agentic-pr-discovery |
| description | Scan open pull requests and reconcile the needs-review label. Use before running agentic-pr-static-review to find PRs needing a review pass. Outputs JSON with reviewed, needs_review, labelled, clean, incomplete, and error arrays. |
This skill is manual-only. It scans open pull requests, including
drafts and pull requests from forks, and reconciles the repository-owned
needs-review label.
The operator must provide a write-permission operator credential manifest.
The repository is read from GITHUB_REPOSITORY unless --repository is
provided. Protected configuration is loaded from .github/agentic-review.
Discovery does not check out branches and does not run tests. It uses the bounded GitHub client through:
python3 -m autoresearch.ar.review.cli discover --operator FILE [--repository OWNER/REPO]
The command prints JSON containing the DiscoverySummary fields
reviewed, needs_review, labelled, clean, incomplete, errors, and
complete. Each item contains a pull request number and reason. It exits
with status 1 when the scan is incomplete.
Use preflight.sh before discovery to validate the credential, protected
configuration, and read/write API boundary without selecting a model or
provider.