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.
npx skills add cmj-hub/claude-evp --skill evp-craftgit clone --depth 1 https://github.com/cmj-hub/claude-evpWrote 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.
[](https://agentmods.dev/skills/cmj-hub/claude-evp/evp-craft)<a href="https://agentmods.dev/skills/cmj-hub/claude-evp/evp-craft"><img src="https://agentmods.dev/badge/skills/cmj-hub/claude-evp/evp-craft/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cmj-hub/claude-evp/evp-craft"><img src="https://agentmods.dev/badge/skills/cmj-hub/claude-evp/evp-craft.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.00986 |
| Opus 5 | $0.00030 | $0.00493 |
| Sonnet 5 | $0.00012 | $0.00197 |
| Haiku 4.5 | $0.00006 | $0.00099 |
Grade A, and why
evp-craft 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.
How it starts
The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
EVP Craft — sub-skill
Generates 3 EVP variants for a specific awareness tier.
Activation
Loaded by evp on:
- "Write an EVP for "
- "EVP for solution-aware prospects"
- "Tier 3 EVP"
Workflow
1. Lock the awareness tier
Ask the user which Schwartz tier (1-5). If they don't know, walk them
through the 5-tier table from ../../evp/SKILL.md and ask: "Which
describes your audience right now?"
Most B2B buyers are in Tiers 2-4. If the user says "Tier 5," push back — that's product-team thinking, not buyer-where-they-are thinking.
2. Capture the inputs
Required:
icp— specific ICP segmentpsp_pain— pain in the prospect's vocabulary (from claude-psp)outcome— specific result you delivertradeoff— obvious tradeoff you spare themproof(optional) — case study / metric backing it
If PSP is missing, suggest the user run claude-psp first. EVPs
without a PSP are guesses.
3. Generate 3 variants
Each variant uses the shape but with different emphasis:
Variant A — outcome-led
"For , in , we're the that does without ."
The outcome carries the line. Use when the outcome is differentiated and concrete (e.g., "ships 14 qualified meetings in 30 days").
Variant B — tradeoff-led
"For , in , we're the team that hits WITHOUT ."
The tradeoff carries the line. Use when the tradeoff is the elephant-in-the-room reason the audience hesitates (e.g., "without hiring 2 more SDRs" or "without sacrificing self-serve").
Variant C — ICP-led
" in use us when they want without ."
The ICP carries the line. Use when the audience is hyper-niche and you want immediate exclusion-by-segment (signals fit fast).
4. Validate each variant
Run the self-check:
| Check | Pass criterion |
|---|---|
| Word count | ≤22 |
| Outcome specificity | A concrete noun + metric, not a verb like "improve" |
| Tradeoff specificity | A concrete thing they'd otherwise have to give up |
| ICP named | Segment is specific enough to exclude wrong-fit |
| Pain vocabulary | Uses their words from the PSP, not category jargon |
| Tier-appropriate | Matches the awareness tier (no Tier-5 line for a Tier-2 audience) |
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.
- yesterday Changed · -1 lines c97215580988
- 10d ago First seen · 140 lines · 59 tokens per session scan A ad430ab59e44
evp-craft is a skill published in the GitHub repository cmj-hub/claude-evp (2 stars, last pushed yesterday), licensed MIT. It adds 59 tokens to every session and 986 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 2026-08-31.
Other skills, from other repositories
go-to-market-playbook
A reusable Go-to-Market strategy template for both B2B and B2C launches. Covers positioning, messaging, ICP definition, channel selection, and competitive analysis frameworks. By @WeiYipei.
psp
Pain Signal Profile (PSP) builder for B2B operators. Translates public, recent, verifiable signals (job posts, funding rounds, leadership changes, product launches, pricing changes, hiring patterns) into the felt operational pain that drives buying decisions — and then into the EVP that speaks to it. Anchors every…
psp-onboarding
First-run interactive setup for the PSP skill pack. Walks the operator through brand-config.json (ICP precision, exclusion criteria, signal sources) and SOUL.md (buyer vocabulary, stories, the 11am-Tuesday moment) in 15 minutes. Refuses to let the operator skip — generic PSP output is worse than no PSP. Loaded…
psp-construct
Step-by-step construction of a Pain Signal Profile for a specific ICP segment. Walks operator through the 5 components (signal, pain, timing, role, vocabulary), validates each, and produces a structured PSP doc. Loaded by the main psp skill when the user wants to build a PSP from scratch.
psp-kickoff
Adaptive router for the PSP skill pack. Detects state (brand-config? SOUL.md? primary PSP drafted? secondary PSPs? vocabulary refreshed?) and picks the next-best step. Loaded by the main psp skill on bare invocation. Inspired by the coldoutboundskills kickoff pattern.
psp-signal-hunt
Surface 5-10 public, recent, verifiable signals worth hunting for a given ICP segment. Catalogs signal types (hiring, funding, leadership, product, pivot, content) and proposes operational sources + cadence to track each. Loaded by the main psp skill when the user wants to find what signals to hunt.