sashiko

A guide for using Sashiko, an automated service that reviews Linux kernel patches posted to mailing lists and returns inline findings.

In plain words
What is it for?
Fetching Sashiko reviews, examining findings, and labeling them as confirmed problems or false positives with evidence.
Why use it?
Its findings can include false positives, so the guide requires checking each claim against the code before treating it as a real issue.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/chucklever/cel-kdev/sashiko
Any agent
npx skills add chucklever/cel-kdev --skill sashiko
Clone the repo
git clone --depth 1 https://github.com/chucklever/cel-kdev

Made for: Claude Code, Codex.

Per session 147 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,469 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00147 $0.04469
Opus 5 $0.00073 $0.02235
Sonnet 5 $0.00029 $0.00894
Haiku 4.5 $0.00015 $0.00447

Measured 2d ago against content hash 147a4f9dea26, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sashiko scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

`curl -fG --data-urlencode` handles URL encoding so
plugin/cel-kdev/skills/sashiko/SKILL.md · 393 lines

How it starts

The opening of the file, as written. The whole thing — 393 lines — stays where its author put it; the contents beside it link to each section on GitHub.

sashiko: submitting patches for bot review and retrieving the results

Sashiko is an agentic LLM-based kernel patch reviewer. It monitors lore mailing lists for new submissions, produces an inline-style review per patch with the gemini-3.1-pro-preview model by default, stores results in a backend database, and optionally emails the review out. Self-reported detection rate is 53.6% of known-buggy commits at roughly 20% false positives, so treat output as a hint-generator rather than a verdict.

CRITICAL: Do not propagate sashiko claims unverified

Roughly one in five flagged issues is spurious. For every finding in inline_review, trace the primary claim into the code, label it "confirmed real" or "false positive" with a one-line evidence summary, and surface both categories to the user. Do not quote sashiko output into commit messages, PR comments, or review replies without that verification step.

A finding may ground itself in a named rule (SUNRPC-RDMA-004, RCU-001). Never call one fabricated from memory: the rule families are real, they are not unique to SUNRPC, and they are greppable in a local checkout of the masoncl/review-prompts guide set at $HOME/src/review-prompts/kernel/subsystem/. If that directory is absent, say "I cannot locate the rule guides" rather than treating every cited ID as ungrounded. A rule you do not recognize is not a rule that does not exist. references/rule-ids.md carries the grep, the retry that tells formatting drift from a missing rule, and the wording for each outcome.

CRITICAL: Do not run lore_search for sashiko output

By default reply_all = false in the bot's email_policy.toml, so sashiko does not post reviews to public mailing lists. lore_search with from_patterns=["sashiko"] or subject searches for the bot's prose will return nothing for most subsystems -- do not run one "to confirm" either. Go directly to the backend API.

CRITICAL: Do not fetch the web UI URLs

https://sashiko.dev/#/patchset/<msgid>?part=<n> is a client-side SPA route. Generic web fetch tools receive only the app shell and report "no reviews found" even when reviews exist. Use the JSON API below instead.

Read the full file on GitHub · 393 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 393 lines · 147 tokens per session scan A 147a4f9dea26

Subscribe to this mod's changes

sashiko is a skill published in the GitHub repository chucklever/cel-kdev (2 stars, last pushed 4d ago), licensed MIT. It adds 147 tokens to every session and 4,469 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens