← 返回 Skills 市场

linux-riscv-contribute

Name: linux-riscv-contribute
Author: zcxggmu

作者 zcxGGmu · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

308

总下载

当前安装

版本数

在 OpenClaw 中安装

/install linux-riscv-contribute

功能描述

Orchestrate an OpenClaw multi-agent pipeline to close Linux RISC-V gaps versus ARM/x86 (Linux tree + KVM lore), create and manage GitHub issues, generate des...

使用说明 (SKILL.md)

Linux RISC-V Contribute

Overview

Use this skill to run a repeatable discover -> issue -> plan -> implement -> patch pipeline with OpenClaw as orchestrator and ACP agents (claude-code, codex) as workers.

Keep humans at exactly three gates:

Confirm gap triage and priorities.
Approve implementation plan.
Approve final patch email before sending.

Workflow

Step 0: Bootstrap workspace

Run scripts/bootstrap_openclaw_workflow.sh \x3Cdocs_repo_root> \x3Clinux_repo_path> to create/update:

kernel/openclaw/config/workflow.yaml
kernel/openclaw/state/{gap_registry.yaml,issue_map.yaml,run_history/}
kernel/openclaw/{plans,patches,logs}

If files already exist, do not overwrite without explicit user approval.

Step 1: Discover RISC-V gaps

Collect evidence from:

Linux source tree (arch/riscv, arch/arm64, arch/x86, virt/kvm)
KVM lore (https://yhbt.net/lore/kvm/)

Write structured entries to state/gap_registry.yaml with:

gap_id, type (feature|performance|maintainability), summary
evidence (paths, commits, lore URLs)
severity (P0|P1|P2), confidence (high|medium|low)
acceptance_hint

Pause for Gate-1 human triage before creating issues.

Step 2: Sync GitHub issues

For each approved gap:

Create/update issue in configured repo.
Add labels from severity/type.
Save gap_id -> issue_number mapping to state/issue_map.yaml.

Use one issue per gap; avoid duplicate issues by matching gap_id.

Step 3: Plan with Claude Code (ACP)

Spawn ACP session explicitly:

runtime: "acp"
agentId: "claude-code"

Ask for:

file-level design
test matrix (kselftest, kvm-unit-tests, perf)
rollback/risk notes
upstreaming strategy

Save outputs under kernel/openclaw/plans/issue-\x3Cid>-plan.md. Pause for Gate-2 human plan approval.

Step 4: Implement and verify with Codex (ACP)

Spawn ACP session explicitly:

runtime: "acp"
agentId: "codex"

Run iterative loop until pass or policy limit:

Implement approved plan.
Build and run configured tests.
Parse failures and patch.

Record each iteration in state/run_history/*.json. If max iterations reached, return to Step 3 with failure summary.

Step 5: Generate patch and email package

Produce:

git format-patch series
checkpatch result
suggested To/Cc (get_maintainer.pl, lore context)
cover letter draft

Save artifacts in kernel/openclaw/patches/. Pause for Gate-3 human send approval.

Only send to mailing lists after explicit approval.

OpenClaw execution rules

Prefer ACP sessions_spawn for agent work; set agentId explicitly.
Limit parallel issues to 2-3 unless user changes policy.
Never auto-send external email without user confirmation.
Preserve auditability: every stage must have file artifacts.

Quick command prompts for operator

Use these ready prompts in OpenClaw chat:

按 workflow.yaml 执行 Step-1，更新 gap_registry.yaml，并生成 Gate-1 审核表。
基于已批准 gap 执行 Step-2，同步 issue 并输出映射表。
对 issue #\x3Cn> 用 claude-code 执行 Step-3，生成详细方案和测试矩阵。
对 issue #\x3Cn> 用 codex 执行 Step-4，直到验证通过或达到迭代上限。
对 issue #\x3Cn> 执行 Step-5，先 dry-run 生成 patch 和发信草案，等待我确认。

References

Workflow template: references/workflow-template.yaml
Issue template: references/issue-template.md
Human gate checklist: references/gate-checklist.md

安全使用建议

This skill appears to do what it says (automate a discover->issue->plan->implement->patch pipeline) but has a few important gaps and risks you should consider before installing: - Authentication: The skill will create/update GitHub issues and prepare/send patch emails, yet it declares no GitHub token or mail/send credentials. Confirm how the platform will supply authentication (agent secrets, user-provided tokens, or manual steps). Do not assume it will work without configuring credentials. - Data exposure: The workflow reads local Linux source trees and instructs spawning external ACP agents (claude-code, codex). That means potentially sensitive repository content and internal design details will be sent to external model runtimes. If your code or metadata must remain private, run this only in an environment where sending data to those models is permitted, or restrict the agent prompts to safe excerpts. - Human gates: The skill includes three explicit human approval gates and a 'never auto-send external email without user confirmation' rule; ensure the platform enforces these gates before you allow any autonomous runs. - Review templates and bootstrap script: The bootstrap script is simple and only writes under the docs repo you pass in, but verify the target path before running and keep backups. Also review workflow-template.yaml for hardcoded repo names/models (e.g., 'zcxGGmu/linux-riscv-docs', 'hanbbq/gpt-5.3-codex') and replace with your own configuration. - Test in dry-run/isolated environment first: Use the workflow's dry_run flags, run on a non-sensitive copy of repos, and confirm what data is transmitted to ACP agents and what the platform requires for GitHub/mailing authentication. If the project owner can clarify how credentials are provided (or update the skill to declare required env vars and data-sanitization rules), the remaining concerns would be addressed. Until then, treat the skill as functional but requiring careful ops/credential review.

功能分析

Type: OpenClaw Skill Name: linux-riscv-contribute Version: 1.0.0 The skill bundle orchestrates a complex workflow for Linux kernel contribution that involves high-risk capabilities such as automated code implementation, building, and testing (RCE by design) via the 'codex' agent. While it incorporates human-in-the-loop 'gates' for approval, the 'scripts/bootstrap_openclaw_workflow.sh' file contains hardcoded references to a specific GitHub repository (zcxGGmu/linux-riscv-docs) and a non-existent model name (gpt-5.3-codex), which could lead to unintended data exposure or unexpected agent behavior if the configuration is not manually audited and corrected by the user.

能力评估

⚠ Purpose & Capability

The skill's stated purpose (discover -> issue -> plan -> implement -> patch) matches the instructions and included files. However, it promises to create/update GitHub issues and prepare/send patch emails while declaring no credentials, tokens, or config paths for GitHub/mailing delivery. That omission is an incoherence unless the platform supplies these credentials implicitly; the SKILL.md does not document that assumption.

⚠ Instruction Scope

Runtime instructions tell the agent to read large code trees (arch/riscv, arch/arm64, arch/x86, virt/kvm) and spawn ACP agent sessions (claude-code, codex) to perform design and implementation. This is expected for the purpose, but it explicitly instructs sending file-level designs and test artifacts to external ACP runtimes — which may transmit repository code and context to third‑party models. The skill does include three human gates and a strict 'no auto-send' rule for email, which reduces but does not eliminate the risk. The SKILL.md does not limit or describe what data is safe to send to ACP agents.

✓ Install Mechanism

No install spec is present (instruction-only plus one bootstrap script). The provided bootstrap script only creates a workflow config and minimal state files under the user-supplied docs repo path and is straightforward and readable (no remote downloads or extraction).

⚠ Credentials

The skill declares no required environment variables or primary credential, yet its operations (GitHub issue creation, optionally sending emails, and calls to external ACP runtimes/models) normally require credentials or platform-level API access. The workflow-template.yaml embeds repo and mailing targets but provides no mechanism for authentication. This mismatch should be resolved or documented (e.g., 'OpenClaw will use agent-level secrets' or declare required env vars).

✓ Persistence & Privilege

The skill is not always-enabled and does not request elevated system presence. Its bootstrap script writes files only under the user-provided docs_repo path and respects existing files (it skips existing workflow.yaml). It does not modify other skills or system-wide settings.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install linux-riscv-contribute
安装完成后，直接呼叫该 Skill 的名称或使用 /linux-riscv-contribute 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of linux-riscv-contribute skill - Automates Linux RISC-V kernel contribution workflow using a human-in-the-loop, multi-agent pipeline. - Supports gap discovery, GitHub issue management, design planning, implementation/testing, and patch email generation. - Integrates OpenClaw as orchestrator with Claude Code and Codex ACP agents. - Gate-based checks require human approval at triage, plan, and patch send steps. - Includes workspace bootstrapping instructions, file artifact auditing policy, and operator command prompts.

元数据

Slug linux-riscv-contribute

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题