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 Xakki/ai-agents-skills --skill model-tiersgit clone --depth 1 https://github.com/Xakki/ai-agents-skillsWrote 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/xakki/ai-agents-skills/model-tiers)<a href="https://agentmods.dev/skills/xakki/ai-agents-skills/model-tiers"><img src="https://agentmods.dev/badge/skills/xakki/ai-agents-skills/model-tiers/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/xakki/ai-agents-skills/model-tiers"><img src="https://agentmods.dev/badge/skills/xakki/ai-agents-skills/model-tiers.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 | $0.00058 | $0.00643 |
| Opus 5 | $0.00029 | $0.00321 |
| Sonnet 5 | $0.00012 | $0.00129 |
| Haiku 4.5 | $0.00006 | $0.00064 |
Grade A, and why
model-tiers 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 11d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
model-tiers — cheap / standard / judgment
Three cognitive tiers — skills and delegation rules name tiers only, never vendor
model slugs. Resolved slugs for this session are injected at SessionStart
(hooks/model-tiers-inject.sh) and appear in context as cheap, standard,
judgment.
Tiers
| Tier | Role | Default slug |
|---|---|---|
| cheap | Mechanical, spec-complete, single-pass work (grep, rename, format, inventory) | haiku |
| standard | Routine product logic, known patterns, doc/skill edits, code review | sonnet |
| judgment | Architecture, tricky debug, cross-file synthesis, multi-source research | opus |
When torn between two tiers, pick the cheaper one; escalate on ambiguity. Floor = standard for anything touching product logic.
Env keys
| Key | Tier |
|---|---|
AI_MODEL_CHEAP |
cheap |
AI_MODEL_STANDARD |
standard |
AI_MODEL_JUDGMENT |
judgment |
Precedence
- Runtime environment —
AI_MODEL_*values inherited when the agent starts. Claude, Codex, and Cursor can set these through theirsettings.jsonenvblock. For Prime Agent, export them in the shell/service that startsprime-agent; use exact selectors accepted byprime-agent model list. - Plugin defaults —
skills/model-tiers/defaults.env(shipped with the plugin).
Unset override keys fall back to defaults. Never invent a model — always use the resolved slug from the map above or from SessionStart injection.
SessionStart inject
Every session receives a tiny block:
## Model tiers (resolved for this session)
cheap = <slug>
standard = <slug>
judgment = <slug>
Pass these slugs as model: on every Agent / Task call. Omitting model: inherits
the caller's tier (expensive default).
Configure overrides
Set AI_MODEL_* in the active runtime's environment. Claude, Codex, and Cursor
can use their settings.json env block; for Prime Agent, export the values
before startup. Long jq examples → reference.md.
What ships with it
3 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.
- 11d ago First seen · 73 lines · 58 tokens per session scan A a031144e5bf9
model-tiers is a skill published in the GitHub repository Xakki/ai-agents-skills (6 stars, last pushed 20d ago), licensed MIT. It adds 58 tokens to every session and 643 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
openpanel
Query self-hosted OpenPanel analytics — funnel counts, unique users, breakdowns by os/country/path, raw events. Use whenever analytics numbers are needed: funnel conversion, visitor counts, event counts over a window, period-over-period deltas, "how many signups/checkouts/sessions", or any /darkflow:analytics-review…
taiyi-evolve
A workflow skill that compares the implemented code with the frozen design after development and testing. It records architecture changes and proposes updates to DESIGN.md, the document describing the intended system structure.
taiyi-diagram-c4
A code-scanning tool that builds C4 architecture documents from a repository. It separates facts observed in the code from conclusions inferred about the design and uses Mermaid diagrams as the source format.
docs
Use when documentation generally needs to catch up with a branch before pushing or merging, covering internal developer docs, external customer-facing docs, and release notes together — "update the docs", "do a docs pass before I merge", "make sure everything's documented". Prefer this when no single documentation…
legal-risk-assessment
Structure legal-risk reviews by issue, rule source, likelihood, impact, evidence, uncertainty, controls, and escalation.
skill-pack-builder
Create, validate, package, and optionally install original KunAgent skills.