cold-learn

A workflow for reading experiment results from FoxReach and proposing changes to `cold.md`. It compares interested-reply rates between variants with a two-proportion z-test and checks bounce and not-interested rates.

In plain words
What is it for?
Use it to analyse FoxReach experiment variants, compare reply rates statistically, review guard conditions, and prepare or commit a `cold.md` change when permitted.
Why use it?
It helps decide whether an experiment result supports changing the documented approach, while checking basic result-quality safeguards.

Skill for Claude CodeCodex

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 skills/concaption/cold-md/cold-learn
Any agent
npx skills add concaption/cold-md --skill cold-learn
Clone the repo
git clone --depth 1 https://github.com/concaption/cold-md

Made for: Claude Code, Codex.

Per session 133 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,995 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin unknown 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 $0.00133 $0.02995
Opus 5 $0.00067 $0.01497
Sonnet 5 $0.00027 $0.00599
Haiku 4.5 $0.00013 $0.00299

Measured 2d ago against content hash b529a6e74c19, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cold-learn scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Same CLI → curl → docs pattern as cold-leads/cold-send. The new `inbox categorize-stats` endpoint:
plugin/cold-md/skills/cold-learn/SKILL.md · 307 lines

The source is not reproduced here

No licence file

A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.

Read it on GitHub

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 · 307 lines · 133 tokens per session scan A b529a6e74c19

Subscribe to this mod's changes

cold-learn is a skill published in the GitHub repository concaption/cold-md (10 stars, last pushed 3mo ago), with no licence file. It adds 133 tokens to every session and 2,995 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

outbound-strategy

Build your own in-house cold email / outbound infrastructure instead of renting Smartlead, Instantly or Lemlist. Use when someone wants to set up self-hosted cold outreach: connecting their own SMTP/IMAP inboxes, sourcing and segmenting B2B leads, writing multi-step sequences, sending at scale with deliverability-safe…

gquthier/autonomous-cold-email · 144 tokens

email-copywriting

Write effective cold email copy using proven principles for hooks, body, CTAs, signatures, AI personalization, and QA processes. Use this skill when (1) writing email sequences from scratch, (2) consuming Campaign Strategy output to write email copy based on an email sequence blueprint and AI variables, or (3)…

Ken-Technology/cold-email-skills · 95 tokens

prompt-writer

Generate AI personalization prompts for cold email campaigns. Reads the AI variables from a campaign strategy and outputs detailed prompts you can run in any personalization tool or LLM to tailor each email per prospect. Use after campaign strategy, or when creating prompts for an existing sequence. Triggers on "write…

Ken-Technology/cold-email-skills · 71 tokens

email-review

Review cold email copy for quality, compliance with critical copy rules, and client preferences. Use after email-copywriting completes to validate and improve email sequences. Scores emails 1-5 and outputs emailsv2.md when score >= 4.5. If score < 4.5, provides revision feedback for email-copywriting to iterate.…

Ken-Technology/cold-email-skills · 96 tokens

campaign-strategy

Create email campaign strategy for a single segment. Generates email sequence blueprints and AI personalization variables for one segment within a campaign plan. Receives a segment folder path and reads the segment's angle from the parent plan.md. Triggers on requests like 'create strategy for segment', or when…

Ken-Technology/cold-email-skills · 67 tokens

lead-magnet

Ideate and design lead magnets for cold email campaigns. Covers both click-based (downloads, tools, signups) and reply-based (free trials, guarantees, done-for-you samples) campaign types. Use when a client needs a new lead magnet, when campaign-planning identifies a weak offer, or when the user wants to brainstorm…

Ken-Technology/cold-email-skills · 115 tokens