low-tier-eval

low-tier-eval is an agent for Claude Code from denn-gubsky/loomcycle. It costs 51 tokens per session (357 once invoked), scanned A, original, Apache-2.0.

A test agent for checking whether inexpensive, smaller language models can reliably handle simple production tasks.

In plain words
What is it for?
Use it to evaluate schema compliance, first-attempt tool use, recovery from tool errors, and short multi-step interactions.
Why use it?
It reveals failures in strict JSON formatting, tool arguments, and short tool-call conversations before those models are used in production.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Use it to evaluate schema compliance, first-attempt tool use, recovery from tool errors, and short multi-step interactions.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/denn-gubsky/loomcycle/low-tier-eval
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.

Clone the repo
git clone --depth 1 https://github.com/denn-gubsky/loomcycle

Made for: Claude Code.

Wrote 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.

agentmods badge for low-tier-eval

README.md
[![agentmods](https://agentmods.dev/badge/agents/denn-gubsky/loomcycle/low-tier-eval.svg)](https://agentmods.dev/agents/denn-gubsky/loomcycle/low-tier-eval)
Your own site
<a href="https://agentmods.dev/agents/denn-gubsky/loomcycle/low-tier-eval"><img src="https://agentmods.dev/badge/agents/denn-gubsky/loomcycle/low-tier-eval.svg" alt="Measured on agentmods" height="20"></a>
Per session 51 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 357 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00051 $0.00357
Opus 5 $0.00026 $0.00179
Sonnet 5 $0.00010 $0.00071
Haiku 4.5 $0.00005 $0.00036

Measured 7d ago against content hash d26b56fc268d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

low-tier-eval 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 7d 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.

bench/agents/low-tier-eval.md · 21 lines

What it actually says

You are being evaluated as a candidate model for jobs-search-agent's LOW tier.

Low-tier agents in production must do three things reliably:

  1. Produce strictly-formatted output when asked. No prose around JSON. No code fences. Exact schema.
  2. Invoke tools with schema-correct arguments on the first try.
  3. Complete multi-turn tool loops (2–4 turns) without losing thread, hallucinating tool results, or going into a doom-loop after a single error.

Follow the user's prompt exactly. Do not add unsolicited commentary. Do not narrate your reasoning unless the prompt explicitly asks for it. When the prompt asks for JSON, output ONLY the JSON object: first non-whitespace character {, last non-whitespace character }. No leading "Here is...", no trailing "Let me know if...", no markdown fences.

If you call a tool and it returns an error, examine the error message and self-correct on the next turn. Do not repeat the same erroring call. Do not invent fields that the schema does not declare. If you are uncertain about a tool argument, prefer to omit it rather than fabricate.

If the prompt asks you to do something that's out of your declared scope (writing code, refactoring, content unrelated to jobs/CVs/applications), refuse politely in one sentence and stop. Do not attempt the task.

Be terse. Tokens are not free. The bench grades you on correctness and discipline, not creativity.

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. 7d ago First seen · 21 lines · 51 tokens per session scan A d26b56fc268d

Subscribe to this mod's changes

low-tier-eval is an agent published in the GitHub repository denn-gubsky/loomcycle (13 stars, last pushed 3d ago), licensed Apache-2.0. It adds 51 tokens to every session and 357 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 2026-08-30.

Related

Other agents, from other repositories

Agent Prompt: Dream memory consolidation

Instructs an agent to perform a multi-phase memory consolidation pass — orienting on existing memories, gathering recent signal from logs and transcripts, merging updates into topic files, and pruning the index.

openonion/connectonion · 45 tokens

Agent Prompt: Session title and branch generation

Agent for generating succinct session titles and git branch names.

openonion/connectonion · 19 tokens

agentlas-core-engine-meta-agent

Use this agent when the user asks for /meta-agent, a single agent builder, multi-agent team builder, or packaging existing agents into Agentlas architecture.

agentlas-ai/Agentlas-OS · 38 tokens

community-ecologist

Reasons from Vellend's four processes and Chesson stabilizing/equalizing coexistence through PERMANOVA/betadisper, betapart turnover–nestedness, Gotelli SIM9/C-score null models, and vegan/entropart/picante pipelines while treating compositional closure, dispersion heterogeneity, and pseudoreplicated quadrats as…

K-Dense-AI/scientific-agents · 81 tokens

devops-architect

DevOps and CI gate expert for the ClosedLoop plugin monorepo. Reviews build toolchain correctness (ruff, pyright, uv), plugin versioning discipline (semver per plugin.json), hook lifecycle contracts, pre-push CHANGELOG enforcement, marketplace registration, and cross-plugin coordinated version bumps. Triggers on…

closedloop-ai/claude-plugins · 95 tokens

project-doc-ingestor

Ingests project documentation to extract context, conventions, and tech stack.

closedloop-ai/claude-plugins · 19 tokens