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 opendatahub-io/ai-helpers --skill adr-reviewgit clone --depth 1 https://github.com/opendatahub-io/ai-helpersWrote 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/opendatahub-io/ai-helpers/adr-review)<a href="https://agentmods.dev/skills/opendatahub-io/ai-helpers/adr-review"><img src="https://agentmods.dev/badge/skills/opendatahub-io/ai-helpers/adr-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/opendatahub-io/ai-helpers/adr-review"><img src="https://agentmods.dev/badge/skills/opendatahub-io/ai-helpers/adr-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 141 Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.Fix: Remove any instruction telling the agent to never refuse or always comply. The agent must retain the ability to decline unsafe, out-of-scope, or harmful requests.
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 | $0.00124 | $0.02429 |
| Opus 5 | $0.00062 | $0.01215 |
| Sonnet 5 | $0.00025 | $0.00486 |
| Haiku 4.5 | $0.00012 | $0.00243 |
Grade B, and why
adr-review scanned grade B 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 12d 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.
Tells the agent never to refusemediumAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
- **Missing sections** — don't refuse to review. Note the gap and continue. Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ADR Review Panel
This skill runs a panel of specialist reviewer subagents over an Architectural Decision Record (ADR) and produces a consolidated report in two formats: a PDF document and a PPTX slide deck.
Why a panel of agents?
Architecture reviews benefit from multiple independent perspectives. A single reviewer tends to anchor on whichever concern they noticed first and under-weight the others. By running specialists in parallel, each focused on one dimension, we get broader, less-biased coverage — and the synthesis step surfaces tensions between perspectives (e.g. "the cheapest option is also the least reversible") that a single reviewer might flatten.
Workflow
Step 1 — Prompt for the ADR location and clarifying context
Always ask the user where the ADR is, even if there's a plausible file in context. Accepted inputs:
- A path to a Markdown file (
.md,.markdown) - A path to a Word document (
.docx) - A path to a directory containing multiple ADRs (review each one)
If the user provides a .docx, extract the text first using the document-skills:docx skill or python-docx. If they provide Markdown, read it directly.
Use the AskUserQuestion tool to gather any clarifying context the reviewers will need. Ask in a single AskUserQuestion call with multiple questions rather than one-at-a-time. Tailor the questions to what is actually unclear after a quick skim of the ADR — don't ask boilerplate. Typical useful questions:
- Audience & stakes — "Who is this review for (author self-check, formal arch board, post-incident retro)?" This shapes tone and severity thresholds.
- Scope — "Are there dimensions you want the panel to emphasize or skip?" (e.g., "skip cost, we already costed it elsewhere")
- Context not in the doc — "Is there background the ADR assumes readers already know? Team size, existing stack, regulatory environment?"
- Decision status — "Is this a draft open to changes, or has the decision already been made and you want a risk audit?"
- Known concerns — "Anything you're already worried about that you want the panel to focus on?"
What ships with it
5 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.
- 12d ago First seen · 154 lines · 124 tokens per session scan B 6c77d7b81dd0
adr-review is a skill published in the GitHub repository opendatahub-io/ai-helpers (37 stars, last pushed 4d ago), licensed Apache-2.0. It adds 124 tokens to every session and 2,429 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it B with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
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.
miru
Use Miru Code Search when the user asks where code lives, how behavior is wired, or what related code paths exist in a repo. Prefer this for conceptual code exploration over grep, glob, or broad file reads.
codex
Delegate coding tasks to the OpenAI Codex CLI for features, refactoring, PR reviews, and batch fixes. Requires the codex CLI and typically a git repository.
code-review
Use when asked to review code. Triggers - "look at this PR", "review this diff", "safe to merge?", change inspection. Input is a diff/branch/PR; output is a severity-ranked list of findings at file:line. If the job is to change code, use development.
code-review
This skill reviews completed implementation before it becomes part of the project's engineering history.