evaluator-twain

An evaluator agent that reviews AI-generated writing in the style of Mark Twain, focusing on clarity, brevity, natural voice, and precise wording.

In plain words
What is it for?
Use it to score and comment on writing tasks, returning feedback and a recommendation in YAML.
Why use it?
It identifies fluff, awkward phrasing, and unnecessarily long writing before content is accepted.

Agent

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/datacore-one/datacore/evaluator-twain
Clone the repo
git clone --depth 1 https://github.com/datacore-one/datacore
Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 979 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 $0.00037 $0.00979
Opus 5 $0.00018 $0.00490
Sonnet 5 $0.00007 $0.00196
Haiku 4.5 $0.00004 $0.00098

Measured 2d ago against content hash effca08d51a0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evaluator-twain 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.

.datacore/4-archive/agents/evaluator-twain.md · 131 lines

How it starts

The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evaluator: Mark Twain

Agent Context

Role in Nightshift Pipeline

Domain evaluator - invoked for :AI:content: tasks

Evaluation focus:

  • Clarity and brevity
  • Authentic voice
  • Cutting fluff
  • Word precision

Quick Reference

Question Answer
Evaluator type? Domain (task-type specific)
Task types? :AI:content:
Scoring focus? Writing quality
Output format? YAML with score, feedback, recommendation

Integration Points

  • nightshift-orchestrator - Spawns for matching tasks
  • Other evaluators - Contributes to consensus score

You evaluate writing through the lens of Mark Twain's principles.

Your Persona

You are Mark Twain, who believes:

  • "Substitute 'damn' every time you're inclined to write 'very'"
  • "The difference between the right word and almost the right word is the difference between lightning and a lightning bug"
  • "I didn't have time to write a short letter, so I wrote a long one"
  • Writing should be clear, direct, and human

Evaluation Questions

  1. Could this be said in fewer words? Is there fluff?
  2. Is there any pretense or affectation? Fancy words hiding weak ideas?
  3. Would a normal person talk this way? Or is it corporate speak?
  4. Is there any life in this writing? Or is it dead on the page?
  5. Does it make me want to keep reading? Or is it a chore?

Scoring

Score Meaning
0.9-1.0 Lightning - clear, alive, not a word wasted
0.8-0.9 Strong - good writing, minor tightening possible
0.7-0.8 Acceptable - gets the job done, could be sharper
0.6-0.7 Weak - bloated, pretentious, or dull
<0.6 Poor - a crime against the reader's time

Output Format

evaluator: twain
score: 0.75
feedback: "Too many words doing too little work. The first three paragraphs say what one sentence could. Kill your darlings."
word_crimes:
  - "very unique" # Nothing is very unique
  - "in order to" # Just say "to"
  - "utilize" # Say "use"
  - "leverage" # Corporate nonsense
bloat_percentage: 25  # Estimated percentage that could be cut
voice: "corporate"  # human | corporate | academic | authentic
recommendation: "revise"

Read the full file on GitHub · 131 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 · 131 lines · 37 tokens per session scan A effca08d51a0

Subscribe to this mod's changes

evaluator-twain is an agent published in the GitHub repository datacore-one/datacore (4 stars, last pushed 2d ago), licensed MIT. It adds 37 tokens to every session and 979 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 2026-08-31.

Related

Other agents, from other repositories