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.
npx skills add xdg/xdg-claude --skill review-walkthroughgit clone --depth 1 https://github.com/xdg/xdg-claudeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xdg/xdg-claude/review-walkthrough)<a href="https://agentmods.dev/skills/xdg/xdg-claude/review-walkthrough"><img src="https://agentmods.dev/badge/skills/xdg/xdg-claude/review-walkthrough/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/xdg/xdg-claude/review-walkthrough"><img src="https://agentmods.dev/badge/skills/xdg/xdg-claude/review-walkthrough.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00098 | $0.02174 |
| Opus 5 | $0.00049 | $0.01087 |
| Sonnet 5 | $0.00020 | $0.00435 |
| Haiku 4.5 | $0.00010 | $0.00217 |
Grade A, and why
review-walkthrough scanned grade A with 0 findings 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 11d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
review-walkthrough
An interactive, one-unit-at-a-time walkthrough of a file or a diff, paced by the user. Claude presents each function, annotates it, and flags findings; the user reads along and interrogates. Progress lives in a checklist file in the repo, so the walkthrough survives /clear and spans sessions.
This is a teaching/interrogation protocol, not an automated review. The human is in the loop on every unit. Never batch units, never run ahead, never summarize the file wholesale.
When to use
Trigger on: "walk me through this file", "walk me through this diff/branch/these changes", "guided review of X", "explain this file function by function", "let's go through auth.go together", "review walkthrough".
Do not use for: "review this PR", "code review", "review my changes", or any request for an automated verdict on a diff. Those go to code-review tooling. If ambiguous, ask.
Unit and target
The target is either a file (or a small, tightly-coupled set of files) or a diff (a branch range, staged changes, a commit, a PR).
- File target: the default unit is the function, walked in file order. When the file's natural structure differs, pick a better unit: sections of a config file, endpoints of an API spec, classes or methods, top-level blocks of a script. A file set gets one checklist, grouped by file.
- Diff target: the unit is the change group — a set of related hunks forming one coherent modification. See "Diff targets" below. Never use raw hunks as units.
Language-agnostic: nothing in the protocol depends on the language.
Setup (first invocation for a target)
- Pick the checklist path. Default
REVIEW.mdat the repo root. If that name is taken or the repo has a docs convention, propose an alternative (e.g.docs/review-<file>.md) and confirm. If a checklist for this target already exists, resume instead (see below). - Identify the units. File target: outline the file (prefer a structure-outlining tool over reading the whole file) and list its units in file order. Diff target: build change groups per "Diff targets" below and pin the diff spec.
- Identify related files for coupling checks. File target: callers, config, sibling packages, external dependencies the file leans on. Diff target: the blast radius — callers of changed symbols, tests covering changed code. List them in the checklist.
- Write the checklist file using the template below. It must contain the protocol itself, so a fresh session can resume from the file alone without this skill loaded.
- Confirm the agenda with the user (unit choice, order, related files — and for a diff, the grouping itself), then start the first unit.
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.
- 11d ago First seen · 127 lines · 98 tokens per session scan A 87d2c459803a
review-walkthrough is a skill published in the GitHub repository xdg/xdg-claude (20 stars, last pushed 4d ago), licensed Apache-2.0. It adds 98 tokens to every session and 2,174 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…