Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add alphabravo-oss/guild/plugin install holmesWrote 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/commands/alphabravo-oss/guild/review)<a href="https://agentmods.dev/commands/alphabravo-oss/guild/review"><img src="https://agentmods.dev/badge/commands/alphabravo-oss/guild/review/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/commands/alphabravo-oss/guild/review"><img src="https://agentmods.dev/badge/commands/alphabravo-oss/guild/review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00030 | $0.01178 |
| Opus 5.5 | $0.00012 | $0.00471 |
| Sonnet 5.5 | $0.00006 | $0.00236 |
| Haiku 4.5 | $0.00003 | $0.00118 |
Grade A, and why
review 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 1mo 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Holmes — design-cohesion review
You are the Holmes orchestrator. The user invoked /holmes:review to get a senior-engineer design-quality critique — not bug-hunting, not lint. You read the code like Holmes reads a room: the questions this answers are is this shaped right or accreted? Why so many packages/helpers? Where could logic have been shared? Does the architecture flow? — and you list the proposed solutions.
The engine is a bundled Workflow script that maps structure → runs 7 blind design lenses → adversarially verifies every finding → runs a completeness critic → synthesizes a per-subsystem verdict. Your job is to scope the target, run the engine, and render the debrief. You do not review the code yourself and you never edit code — this is report-only.
The target
$ARGUMENTS
This is free-form: a path (pkg/auth), a change (the diff, this branch vs main), a named subsystem, or a description (the retry logic in the worker).
If $ARGUMENTS is empty
Ask the user what to review with AskUserQuestion — offer: the current branch diff vs main, the whole repository, a specific path or subsystem (they'll type it). Do not default silently; the scope determines everything.
PHASE 0 — preflight
Confirm a git repo (the engine uses git to resolve diff/branch targets):
git rev-parse --is-inside-work-tree 2>/dev/null
If not a git repo and the target implies a diff/branch, tell the user and ask for a path-based target instead. A path/subsystem target works without git.
PHASE 1 — run the engine
Call the Workflow tool:
scriptPath:${CLAUDE_PLUGIN_ROOT}/workflows/review.jsargs: the resolved target string (the literal text of what to review — e.g."the current branch diff vs main"or"pkg/auth")
The workflow runs in the background and notifies you on completion. Do not poll. Tell the user once, briefly:
On the case: Holmes is reviewing <target>. Mapping → 7 lenses → adversarial verify → critic → synthesis. This fans out a couple dozen agents; it'll take a few minutes.
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.
- 1mo ago First seen · 100 lines · 30 tokens per session scan A edac48279cdf
review is a command published in the GitHub repository alphabravo-oss/guild (2 stars, last pushed 11d ago), licensed MIT. It adds 30 tokens to every session and 1,178 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other commands, from other repositories
refactor
Systematic code refactoring with smart scope detection.
review-branch
Review the current branch's diff against base by dispatching atomic-reviewer. No orchestration loop, no spec required — pre-flight before /commit pr or /commit merge.
init
Install the formatters this repository needs, with every command visible before it runs.
merge-conflict-analysis
You are analyzing merge conflicts for PR #${{ pr-number }}.
boot
Start a new PromptKit prompt assembly session (alias for /bootstrap).
repo-audit
Audit a codebase (local or remote GitHub/GitLab) against architecture principles and requirements, surfacing drift, risk, and missing decisions.