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 bendrucker/claude/plugin install reviewWrote 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/bendrucker/claude/human)<a href="https://agentmods.dev/skills/bendrucker/claude/human"><img src="https://agentmods.dev/badge/skills/bendrucker/claude/human/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/bendrucker/claude/human"><img src="https://agentmods.dev/badge/skills/bendrucker/claude/human.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 | $0.00070 | $0.01210 |
| Opus 5.5 | $0.00028 | $0.00484 |
| Sonnet 5.5 | $0.00014 | $0.00242 |
| Haiku 4.5 | $0.00007 | $0.00121 |
Grade A, and why
review:human scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Human Review
Put the change in front of Ben and act on what comes back until he approves.
Context
- herdr pane: !
[ "$HERDR_ENV" = 1 ] && printf %s "${HERDR_PANE_ID:-none}" || printf none - Rex block: !
printf %s "${REX_BLOCK:-none}"
Routing
The host is herdr when the herdr pane is set, else Rex when the Rex block is set, else none. attention.ts open reads the same environment and picks the surface itself.
| Host | Diff | --doc <file> |
|---|---|---|
| herdr | reviewr split. Follow Request | plannotator-tui split. Follow Request |
| Rex | tuicr split. Follow Request | plannotator-tui split, gated. Follow Rex Document Gate |
none, or --browser |
Follow Browser Gate | Follow Browser Gate |
In a cloud session, where Plannotator's browser can't reach Ben, name what to review and end the turn instead.
Precondition
Request after the mechanical passes have run. /ship orders them. Standalone on an unreviewed diff, say so in the --summary text rather than running them here.
Request
bun ${CLAUDE_PLUGIN_ROOT}/scripts/attention.ts raise --kind <kind> --summary "<repo> <branch>: <what to review>". The kind iscodefor the diff anddocfor a document, unless the caller passes a more specific--kind(plan,pr-body). A caller's--kindand--summarypass through verbatim. Outside herdr it is a no-op.bun ${CLAUDE_PLUGIN_ROOT}/scripts/attention.ts open --diff, adding--base <ref>when the caller resolved one, oropen --doc <file>for a document. Under herdr,open --diffre-uses a reviewr sidebar already up over this working tree, andopen --docreloads the plannotator-tui pane it opened earlier for the same file, keeping Ben's annotations.- End the turn with one line naming the surface and saying Send resumes the work, with no recap of the change. reviewr and plannotator-tui deliver into this pane once it is idle, so Send lands only after the turn ends.
Resume
With herdr, the next user turn holds the comment batch plus Ben's own remarks. Each block is a path:lines header, the verbatim snippet, and his text, as the herdr plugin's skills/herdr/references/reviewr.md details. Act on all of it. Sending cleared reviewr's store, so the batch in the turn is the only copy. This plugin's UserPromptSubmit hook has already cleared the pane label.
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.
- 4d ago Changed · +2 lines 94fbe00b38f4
- 10d ago Changed 728c51741d95
- 11d ago Changed · +9 lines 0df6e4be84e9
- 21d ago Changed · -1 tokens per session ef48814bb678
- 24d ago Changed e2904cabf925
- 25d ago Changed 72c99b904bae
- 26d ago Changed b9f4917f4d9a
- 28d ago Changed 593d6d4f2517
- 2mo ago First seen · 55 lines · 71 tokens per session scan A 5efbfeeb34ae
review:human is a skill published in the GitHub repository bendrucker/claude (18 stars, last pushed today), licensed MIT. It adds 70 tokens to every session and 1,210 once invoked, about $0.0003 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 2026-09-11.
review:human compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| review:humanbendrucker/claude | Skill | 18 | today | MIT | A | 70 | Claude Code |
| reviewSethGammon/Citadel | Skill | 928 | 4d ago | MIT | A | 17 | Claude Code |
| critical-code-reviewerposit-dev/skills | Skill | 561 | 7d ago | MIT | A | 67 | Claude Code |
| brooks-sweephyhmrright/brooks-lint | Skill | 1.6k | today | MIT | A | 178 | Claude Code |
| pr-reviewnitinjain999/platform-skills | Skill | 71 | yesterday | Apache-2.0 | A | 81 | Claude Code |
| skill-distilleriliaal/whetstone | Skill | 36 | today | MIT | A | 91 | Claude Code |
| review-depthjszmajda/lid | Skill | 118 | today | MIT | A | 93 | Claude Code |