metric-lock

metric-lock is a skill for Claude Code, Codex from EvoClaw/amplify. It costs 35 tokens per session (634 once invoked), scanned A, original, MIT.

A rule that freezes confirmed evaluation choices in a research project. Evaluation choices include the measures, datasets, data splits, random seeds, testing procedure, and comparison methods.

In plain words
What is it for?
It helps keep confirmed research plans unchanged and requires a documented explanation, justification, and evidence before a locked choice can be altered.
Why use it?
It prevents changing the goal or test conditions after seeing results, which could make an experiment look better without a fair reason.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit It helps keep confirmed research plans unchanged and requires a documented explanation, justification, and evidence before a locked choice can be altered.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/evoclaw/amplify/metric-lock
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.

Any agent
npx skills add EvoClaw/amplify --skill metric-lock
Clone the repo
git clone --depth 1 https://github.com/EvoClaw/amplify

Made for: Claude Code, Codex.

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 metric-lock

README.md
[![agentmods](https://agentmods.dev/badge/skills/evoclaw/amplify/metric-lock.svg)](https://agentmods.dev/skills/evoclaw/amplify/metric-lock)
Your own site
<a href="https://agentmods.dev/skills/evoclaw/amplify/metric-lock"><img src="https://agentmods.dev/badge/skills/evoclaw/amplify/metric-lock.svg" alt="Measured on agentmods" height="20"></a>
Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 634 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 $0.00035 $0.00634
Opus 5 $0.00017 $0.00317
Sonnet 5 $0.00007 $0.00127
Haiku 4.5 $0.00003 $0.00063

Measured 7d ago against content hash 1586fda27804, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

metric-lock 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.

skills/metric-lock/SKILL.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Metric Lock (Discipline Layer)

Overview

Changing evaluation metrics after confirmation is not optimization — it is goalpost-moving. This skill activates after plan freeze (G2) and remains in effect until project end.

Core principle: Locked means locked. No silent modifications.

Violating the letter of this rule is violating the spirit of this rule.

The Iron Law

EVALUATION METRICS, ONCE CONFIRMED, ARE IMMUTABLE WITHOUT USER PERMISSION

What Is Locked

After G2 plan freeze, the following are immutable:

  • Primary metrics (e.g., F1, BLEU, AUC)
  • Evaluation protocol (e.g., k-fold, held-out test, leave-one-out)
  • Datasets and splits (train/val/test partitions, exact files)
  • Random seeds
  • Baseline list and their configurations

What Is Allowed

Secondary/supplementary metrics may be added for additional analysis. They cannot replace primary metrics and cannot be used to override primary metric conclusions.

Change Request Process

IF you believe a locked item must change:

1. EXPLAIN: What exactly needs to change?
2. JUSTIFY: Why is the current version flawed?
3. PROVE: Show evidence the locked version is defective (not just suboptimal)
4. WAIT: User decides — not you
5. LOG: Record old value, new value, reason in change_log

Skip any step = unauthorized modification = academic misconduct

Red Flags — STOP

  • Quietly changing evaluation script parameters
  • Reporting a different metric than the one locked
  • Weakening baseline configuration after seeing results
  • Using a different data split than specified
  • Rounding, clipping, or post-processing metrics differently than protocol
  • "Forgetting" a baseline that was performing well

Rationalization Prevention

Excuse Reality
"This metric isn't quite right" You argued for it in Phase 3. To change, go through the change request process.
"A different metric shows better results" That's exactly why changing is forbidden.
"The community now prefers X" If true, convince the user. Don't change silently.
"Just adding a supplementary metric" Adding secondary is fine. Replacing primary is not.
"The baseline used a different metric" Adapt the baseline, don't change your protocol.
"It's a minor adjustment" Minor adjustments to locked items require user approval. Always.
"The metric has a known limitation" All metrics do. Document the limitation; don't swap the metric.

Read the full file on GitHub · 78 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. 7d ago First seen · 78 lines · 35 tokens per session scan A 1586fda27804

Subscribe to this mod's changes

metric-lock is a skill published in the GitHub repository EvoClaw/amplify (12 stars, last pushed 6mo ago), licensed MIT. It adds 35 tokens to every session and 634 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 skills, from other repositories

officecli-word-form

Use this skill to create fillable Word forms (.docx) with real Content Controls (SDT) + legacy FormField checkboxes + MERGEFIELD mail-merge placeholders + document protection. Trigger on: 'fillable form', 'form fields', 'content controls', 'SDT', 'word form', 'fill in', 'only editable fields', 'protect document'…

iOfficeAI/OfficeCLI · 224 tokens

dingtalk_channel_connect

Use a headed browser to automatically complete DingTalk channel integration for QwenPaw. Applicable when the user mentions DingTalk, developer console, Client ID, Client Secret, bot, Stream mode, binding or configuring a channel. Supports pausing when a login page is detected and resuming after the user logs in.

agentscope-ai/QwenPaw · 69 tokens

officecli-data-dashboard

Use this skill to build a multi-element Excel dashboard — Dashboard sheet on open, multiple formula-driven KPI cards, multiple charts, sparklines, and conditional formatting — from CSV or tabular input. Trigger on: 'dashboard', 'KPI dashboard', 'analytics dashboard', 'executive dashboard', 'metrics dashboard', 'CSV to…

iOfficeAI/OfficeCLI · 157 tokens

pdf

A set of instructions for working with PDF files, which are documents designed to preserve their layout across devices.

agentscope-ai/QwenPaw · 95 tokens

make_plan

For external plan request scenarios, guides the Agent to request a clear, actionable, step-by-step plan from a stronger Agent via listagents and chatwithagent, emphasizing that the plan is executed by the requester, not by the consulted Agent.

agentscope-ai/QwenPaw · 51 tokens

gpt-image-2

A skill for generating or editing images with GPT Image 2 across local, host-provided, or advisory setups.

ConardLi/garden-skills · 177 tokens