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 NachoLafuente/5050-gtm --skill disco-prepgit clone --depth 1 https://github.com/NachoLafuente/5050-gtmWrote 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/nacholafuente/5050-gtm/disco-prep)<a href="https://agentmods.dev/skills/nacholafuente/5050-gtm/disco-prep"><img src="https://agentmods.dev/badge/skills/nacholafuente/5050-gtm/disco-prep/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/nacholafuente/5050-gtm/disco-prep"><img src="https://agentmods.dev/badge/skills/nacholafuente/5050-gtm/disco-prep.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 | $0.00116 | $0.01079 |
| Opus 5.5 | $0.00046 | $0.00432 |
| Sonnet 5.5 | $0.00023 | $0.00216 |
| Haiku 4.5 | $0.00012 | $0.00108 |
Grade A, and why
disco-prep 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 2mo 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.
How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Discovery Call Prep
Generate a pre-discovery-call prep brief from an intro email, referral, or company name. The deliverable is markdown in the chat, not a file. The goal: walk into the call already knowing who they are, what to ask, and which past win to drop as proof.
This is a scan doc, not a report. Target output: under 350 words. If you're padding, cut.
Input
The user provides some combination of: an intro email, a company name + domain, prior notes, a LinkedIn profile, or a vague "call with X tomorrow."
Extract: contact name, company, stakeholders on the call, referrer, stated needs, tools mentioned, numbers (size, volume).
If you don't have a company domain, ask before researching. Do not hallucinate.
Pre-Flight (before writing anything)
- Research. Web-search the domain + recent news (last 6 months). Fetch the homepage. Look up named stakeholders.
- Match to past work. Skim your own notes/CRM for analogous clients to anchor the proof in Section 3. Keep a short list of past wins by vertical wherever you keep it.
Voice & Quality Bar
- Specific, not generic. Every question references THIS prospect's context. No "what's your tech stack?"
- Diagnostic, not polite. Each question shapes scope.
- Cut ruthlessly. A disco call is 30 to 45 min. Pick the questions that matter most.
- Short sentences. No fluff. Don't restate what the email already said.
Do not invent
- Only state facts that appear in the source material (email, notes, research).
- Do not infer timelines, deal volume, or team size unless explicitly stated.
- "Eventually" / "down the line" / "future" = NOT a commitment. Treat as parking-lot.
- If the source is silent on something, it becomes a must-ask question, not a claim.
- When tempted to extrapolate ("this means they probably want X by Y"), stop and turn it into a question instead.
Output Format
Exactly 3 sections. Use ## N. Section Name headers. Be terse.
Header block (above section 1)
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.
- 2mo ago First seen · 87 lines · 116 tokens per session scan A 1e061e4d8501
disco-prep is a skill published in the GitHub repository NachoLafuente/5050-gtm (3 stars, last pushed 4mo ago), licensed MIT. It adds 116 tokens to every session and 1,079 once invoked, about $0.0005 per session on Opus 5.5. 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.
disco-prep compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| disco-prepNachoLafuente/5050-gtm | Skill | 3 | 4mo ago | MIT | A | 116 | Claude Code, Codex |
| gt-email-infraGrowth-Today/claude-skills | Skill | 6 | yesterday | MIT | A | 232 | Claude Code, Codex |
| gt-linkedin-contentGrowth-Today/claude-skills | Skill | 6 | yesterday | MIT | A | 228 | Claude Code, Codex |
| gt-linkedin-outbounderGrowth-Today/claude-skills | Skill | 6 | yesterday | MIT | A | 236 | Claude Code, Codex |
| email-infra-bounce-auditGrowth-Today/claude-skills | Skill | 6 | yesterday | MIT | A | 157 | Claude Code |
| gt-cold-email-writerGrowth-Today/claude-skills | Skill | 6 | yesterday | MIT | A | 181 | Claude Code, Codex |
| gt-gtm-strategy-advisorGrowth-Today/claude-skills | Skill | 6 | yesterday | MIT | A | 242 | Claude Code, Codex |