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 olehsvyrydov/AI-development-team --skill review-tiergit clone --depth 1 https://github.com/olehsvyrydov/AI-development-teamWrote 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/olehsvyrydov/ai-development-team/review-tier)<a href="https://agentmods.dev/skills/olehsvyrydov/ai-development-team/review-tier"><img src="https://agentmods.dev/badge/skills/olehsvyrydov/ai-development-team/review-tier/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/olehsvyrydov/ai-development-team/review-tier"><img src="https://agentmods.dev/badge/skills/olehsvyrydov/ai-development-team/review-tier.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 | $0.00118 | $0.01989 |
| Opus 5.5 | $0.00047 | $0.00796 |
| Sonnet 5.5 | $0.00024 | $0.00398 |
| Haiku 4.5 | $0.00012 | $0.00199 |
Grade A, and why
review-tier 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 today.
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 — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review tiering — spend the review where it earns its cost
A full multi-agent review is the strongest defect-finder available and the most expensive thing in the pipeline (measured: ~34 agents / ~1.8M tokens / ~16 min per round). Running it on everything is unaffordable; running it on nothing loses the only check that catches cross-cutting consequences invisible at the edit site. This skill decides how much to spend.
Step 0 — free gates first, always
Refuse to start a review until these are green. They cost nothing and a review of a broken branch is pure waste (this has happened: a full round ran against a branch whose test compile was broken).
- the project's build + full test task
- static-analysis / quality gate, if the project has one
- frontend typecheck and production build, if there is a frontend
Read the tool's own exit code, not a pipeline's, and read the output file rather than a summary. cmd > log; echo $? inside a longer chain reports the last command's status, not the one you care about.
Step 1 — classify the diff (deterministic, never judgement)
Look at changed paths and changed symbols, not at your opinion of the change. Any single match promotes the whole diff to FULL.
FULL — a high-consequence surface is touched:
- destructive operations:
delete,purge,drop,truncate,cascade,revoke, bulk/batch removal - authn/authz: login, session, token, password, capability, role, permission, grant, share
- schema migrations, and any FK with
ON DELETE - money and cost: pricing, billing, budget, quota, rate limit
- external egress: which provider/host/endpoint data is sent to
- a port/interface contract in a shared core module
- transaction boundaries, concurrency, locking
TARGETED — ordinary feature code. Scope finders to the diff plus its callers and callees. This caveat is load-bearing: the review's value is finding the break two files away, so a diff-only scope degrades it into a linter.
Also include the writers of any persistent field the diff reads but does not write. Callers-and-callees follows the call graph; a defect in what a column means propagates along the schema instead, and reads and writes of a column are not connected by any call. A feature that only reads a column can therefore never reach, by any call-graph scoping, the code that decides what it means. Grep the entity field and the raw column name; if there is more than one writer, put those writers in front of the finders and treat the column's semantics as part of the blast radius.
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.
- today First seen · 92 lines · 118 tokens per session scan A 96c1879b1166
review-tier is a skill published in the GitHub repository olehsvyrydov/AI-development-team (17 stars, last pushed today), licensed MIT. It adds 118 tokens to every session and 1,989 once invoked, about $0.0005 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 2026-10-19.
Other skills, from other repositories
review
Use when a change is built and needs review before its PR, when verify runs and specclaw-review is installed, or when the user asks to review, re-review or run the review loop locally.
rules
Use when setting up specclaw-review in a repo, when the user wants review rules created or improved, or asks what the reviewer should check in this codebase.
review-spec
A workflow for comparing an existing feature specification with the project's review checklist and implementation. It looks for missing requirements, differences between documented screens and actual screens, and lessons from rejected submissions.
next-steps
Have the Product Owner survey the project's current state — shipped features, in-flight work, open findings, standing goals — and propose one concrete next feature ready to hand to /story-time or /spark. Use when the user has no idea in hand and asks "what should we build next", when a loop just closed and the next…
increment
Start the Act phase of the SPARK loop: build the increment strictly following the approved plan, task by task, with progress tracked in the plan. Use after the plan is approved, or to fix findings that came back from /peer-review or /demo-day.
fresh-eyes
Audit a finished diff with fresh eyes. The mechanism is double-blind reconciliation: a subagent that never saw the conversation reads the diff blind and reports what it thinks the change does. The main context — which holds the real intent — compares that blind read against what the work was meant to do. The…