evaluate

A command for checking whether research claims are supported by evidence during a Wheat sprint. It compares claims with benchmarks, calculations, prototypes, production data, and related claims.

In plain words
What is it for?
Use it to compile claims, run evaluations, update evidence levels, mark replaced claims, and create an HTML comparison dashboard with tables, charts, and metrics.
Why use it?
It helps resolve conflicting or weak claims instead of leaving decisions based on assumptions. It also identifies topics where evidence is missing.

Command

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 commands/grainulation/wheat/evaluate
Clone the repo
git clone --depth 1 https://github.com/grainulation/wheat
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 567 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 $0.00000 $0.00567
Opus 5 $0.00000 $0.00283
Sonnet 5 $0.00000 $0.00113
Haiku 4.5 $0.00000 $0.00057

Measured yesterday against content hash e401cf33198b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evaluate 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 yesterday.

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.

templates/commands/evaluate.md · 85 lines

How it starts

The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/evaluate — Test claims against reality, resolve conflicts

You are running the evaluation phase of the current Wheat sprint. This is the honesty phase — where claims meet data.

Read CLAUDE.md for sprint context and claims.json for all existing claims.

Process

  1. Run the compiler first:

    npx @grainulation/wheat compile --summary
    

    Identify conflicts, weak evidence areas, and coverage gaps.

  2. Evaluate systematically: For each topic with weak or conflicting evidence:

    • Run benchmarks, cost calculations, feature comparisons
    • Test prototypes against real conditions
    • Pull production metrics from connected tools if available
    • Cross-reference claims against each other
  3. Resolve conflicts: When evaluation produces a clear answer:

    • Update the winning claim's evidence tier
    • Mark the losing claim as superseded with resolved_by
    • Add new evaluation claims (evidence: "tested") if needed
  4. Generate comparison dashboard: Create evidence/<topic-or-sprint-slug>.html — a dashboard-style HTML page showing:

    • Side-by-side comparisons with real numbers
    • Conflict resolutions with evidence
    • Data tables, charts (CSS-only), metrics

Claim updates

Evaluation claims use evidence tier tested or production:

{
  "id": "e001",
  "type": "factual",
  "topic": "<what was evaluated>",
  "content": "<measured result — always include numbers>",
  "source": {
    "origin": "evaluation",
    "artifact": "evidence/<slug>.html",
    "connector": null
  },
  "evidence": "tested",
  "status": "active",
  "phase_added": "evaluate",
  "timestamp": "<ISO timestamp>",
  "conflicts_with": [],
  "resolved_by": null,
  "tags": []
}

Update meta.phase to "evaluate" in claims.json.

Run the compiler again

npx @grainulation/wheat compile --summary

Verify that conflicts are resolved. If the compiler still shows blockers, tell the user what remains.

Git commit

Stage claims.json and all new/changed files.

Read the full file on GitHub · 85 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. yesterday First seen · 85 lines · 0 tokens per session scan A e401cf33198b

Subscribe to this mod's changes

evaluate is a command published in the GitHub repository grainulation/wheat (20 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 567 tokens. 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.