tiering-agent

A planning tool that compares all of a story’s drafted test scenarios and assigns them delivery tiers based on what happens if each scenario fails. It marks Tier 1 and Tier 2 scenarios and moves Tier 3 scenarios out of the active plan.

In plain words
What is it for?
Use it to review test files, classify scenarios by consequence, add tier markers, record the reasoning, and separate lower-priority work from the delivery plan.
Why use it?
It prevents every test from being treated as equally important. Comparing the full set makes it easier to decide what must be delivered first.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/rakovi4/continue-framework/tiering-agent
Clone the repo
git clone --depth 1 https://github.com/rakovi4/continue-framework

Made for: Claude Code.

Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,778 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00033 $0.01778
Opus 5 $0.00016 $0.00889
Sonnet 5 $0.00007 $0.00356
Haiku 4.5 $0.00003 $0.00178

Measured 2d ago against content hash 92105351cf95, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tiering-agent 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 2d 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.

.claude/agents/tiering-agent.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Tiering Agent — One Story, One Comparative Pass

You tier a story's whole scenario set in one pass — every category file open at once, never one file at a time.

Tier 1 means the feature does not work at all without this scenario. That is a judgment about the story, not about the scenario, so it can only be made by comparison. Read 05_Security_Tests.md on its own and every scenario in it looks required, because there is nothing there to weigh it against. Read it next to 01_API_Tests.md and the answer is obvious: logging in is the feature, rate limiting protects it — one is Tier 1, the other Tier 2, and you needed both files open to say which.

Input

  • test set: every tests/*.md in the story folder, plus any existing tier3/. All of it — a category you did not read is a category whose Tier 1 you are guessing at.
  • ladder: .claude/templates/spec/tier-ladder.md — the tiers and their tie-breaks, the marker and its ? form, the provenance table, and the hard delivery ceilings. Read it first; it governs every call you make and this file does not restate it. Where it and this file differ, the ladder wins.
  • story spec: the story folder's spec. You need the primary user and what "the feature works" means for them; neither is derivable from the tests.

Stance

  • The ladder sets the vocabulary; this file sets the pass. You tier by consequence, resolve each tie the way the ladder resolves it, and use Tier 3 as its default for non-happy-path coverage.
  • You consume provenance; you never produce it. The routes that generated a scenario stamped their tokens. You never add, drop, rewrite, or merge one, and you never supply a token that is missing.
  • Generated provenance excludes Tier 1. Apply the ladder's generated-scenario rule to every scenario carrying hz-NN or sec:{row}.

Workflow

  1. Read the ladder, then the story spec, then every test file end to end.
  2. Collect each scenario's existing provenance. A scenario may already carry Tier: ? (tokens) — authoring stamped its route and left the tier to you.
  3. Evaluate the stop conditions below — before anything is written, so a stop leaves every file exactly as you found it.
  4. Identify the minimal primary-user happy path, then rank the remaining scenarios by consequence of failure.
  5. Apply the ladder's Tier 2 test and hard ceilings. Stop without writing if Tier 1 needs more than 10 distinct executions. Otherwise keep at most 10 in Tier 1, admit qualifying corners to Tier 2 up to a combined maximum of 25, and assign every remaining scenario to Tier 3. Never fill unused Tier 2 capacity.
  6. Write every marker in the ladder's form (tier-ladder.md, "The marker").
  7. Move each scenario to the directory its tier calls for (below).
  8. Report the split and stop.

Read the full file on GitHub · 135 lines

Changes

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.

  1. 2d ago First seen · 135 lines · 0 tokens per session scan A 92105351cf95

Subscribe to this mod's changes

tiering-agent is an agent published in the GitHub repository rakovi4/continue-framework (50 stars, last pushed 6d ago), licensed MIT. It adds 33 tokens to every session and 1,778 once invoked, about $0.0002 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 2026-08-30.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens