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 grimoire-rs/grimoire --skill hex-reviewgit clone --depth 1 https://github.com/grimoire-rs/grimoireWrote 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/grimoire-rs/grimoire/hex-review)<a href="https://agentmods.dev/skills/grimoire-rs/grimoire/hex-review"><img src="https://agentmods.dev/badge/skills/grimoire-rs/grimoire/hex-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/skills/grimoire-rs/grimoire/hex-review"><img src="https://agentmods.dev/badge/skills/grimoire-rs/grimoire/hex-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 | $0.00182 | $0.07217 |
| Opus 5 | $0.00091 | $0.03608 |
| Sonnet 5 | $0.00036 | $0.01443 |
| Haiku 4.5 | $0.00018 | $0.00722 |
Grade A, and why
hex-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 5d 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 — 530 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hex-review — Adversarial Reviewer
Thin dispatcher. It resolves the review target and baseline, classifies the
diff's tier, resolves overlays, runs the single meta-plan approval gate,
announces the resolved config, and hands off to the matching tier file.
Unlike hex-plan and hex-execute, hex-review never edits its own output
or loops on it — the reviewer panel runs once per invocation and
reports; iterating on the findings is
/hex-execute's job.
The phase plans live in the five tier-<tier>.md files; the shared vocabulary (tiers, worker roles, model classes,
the memory file) lives in the hex-core reference library and is linked
here, never copied.
Shared contracts:
protocol.md ·
workers.md ·
models.md ·
memory.md.
If hex-core is not installed: grim add ghcr.io/michael-herwig/arcana/hex-core:latest.
Argument syntax
/hex-review [tier] <target> [flags]
- tier (optional):
low | medium | high | xhigh | max | auto. Defaultauto— the classifier pickslow…xhighfrom diff metrics.maxis explicit only —--tier=max, or a plan whose Status block saysTier: max; the classifier never emits it (protocol.md). - target (one of):
- empty — the working tree: uncommitted changes (staged + unstaged)
against
HEAD. If the working tree is clean, fall back to the current branch's diff against the resolved baseline and say so in the announce block. - a branch name — diffed against the resolved baseline.
<N>/#<N>/PR <N>/pull/<N>/ a full GitHub PR URL — an explicit pull request.- a path to a markdown plan, ADR, or spec artifact — review the file
directly, not a code diff. Diff metrics do not apply; see
classify.md.
- empty — the working tree: uncommitted changes (staged + unstaged)
against
- flags (before the target, by convention):
--base=<git-ref>— the diff baseline. Pipeline input, not an overlay axis: everything downstream (metrics, tier, review scope, adversary scope) reads from it. Default resolution in step 2.--breadth=minimal|full|adversarial— override the Stage 2 panel breadth (seeoverlays.md).--rca=on|off— override root-cause (Five Whys) analysis.--adversary/--no-adversary— force the cross-model pass on or off.--dry-run— make the meta-plan gate block for explicit approval and stop there (preview only, no workers launched).
What ships with it
7 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.
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.
- 5d ago First seen · 530 lines · 182 tokens per session scan A f8ce8613b977
hex-review is a skill published in the GitHub repository grimoire-rs/grimoire (10 stars, last pushed yesterday), licensed Apache-2.0. It adds 182 tokens to every session and 7,217 once invoked, about $0.0009 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 2026-09-08.
Other skills, from other repositories
code-review-excellence
Master effective code review practices to provide constructive feedback, catch bugs early, and foster knowledge sharing while maintaining team morale. Use when reviewing pull requests, establishing review standards, or mentoring developers.
multi-reviewer-patterns
Coordinate parallel code reviews across multiple quality dimensions with finding deduplication, severity calibration, and consolidated reporting. Use this skill when organizing multi-reviewer code reviews, calibrating finding severity, or consolidating review results.
oracle
Oracle second-model review: bundle prompts/files, debug, refactor, design.
pr-review
Review open PRs two ways - default is a per-PR deep review with severity-tagged findings, inline comments, and a verdict; --survey runs a risk-tiered triage digest of what's safe to merge first.
pr-triage
First-touch triage for external pull requests - verdict, label, and a welcoming comment within minutes of open.
[REPLACE: SKILL_NAME]
First-touch review of newly opened PRs on [REPLACE: WATCHEDREPO] — verdict + welcoming comment + label.