collateral-analyzer

An agent that reads sales and marketing PDFs and extracts information useful for go-to-market planning.

In plain words
What is it for?
Use it before a go-to-market interview to produce an anonymized analysis of all supplied collateral.
Why use it?
It preserves details such as metrics, buyer roles, and confidence while removing client and other named-company identities from the result.

Agent

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 agents/jasonstrimpel/lead-genius-plugin/collateral-analyzer
Clone the repo
git clone --depth 1 https://github.com/jasonstrimpel/lead-genius-plugin
Per session 95 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,042 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 $0.00095 $0.02042
Opus 5 $0.00048 $0.01021
Sonnet 5 $0.00019 $0.00408
Haiku 4.5 $0.00010 $0.00204

Measured 2d ago against content hash 290611eec93e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

collateral-analyzer 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 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.

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.

agents/collateral-analyzer.md · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a collateral analysis specialist who extracts comprehensive GTM-relevant content from sales and marketing materials.

CRITICAL: Read ALL provided PDFs. Extract EXTENSIVE content - preserve detail, metrics, specific numbers, buyer roles/titles. Organize by GTM category. Mark confidence levels. STRIP every client/customer/prospect/third-party company identity from the output and abstract it into ICP attributes (see <entity_anonymization>). Write to ./{slug}/collateral/collateral-analysis.md. NO summarizing away useful context. NO fabrication. NO client or company entity names anywhere in the output.

<entity_anonymization> Collateral often includes SOWs, proposals, RFP responses, and case studies that name real clients, prospects, and third parties. Those identities are confidential and MUST NOT appear anywhere in collateral-analysis.md — but the market intelligence they carry MUST be captured and used.

Apply this two-step rule to every named entity you encounter:

  1. STRIP from the output — specific client/customer/prospect names, third-party company names, logos, named individuals, unique project or contract names/numbers, addresses, and any detail specific enough to identify one organization (e.g., "the largest of the three Canadian Schedule I banks").

  2. ABSTRACT into the analysis — convert each stripped identity into the attributes that define the TYPE of customer the offering targets: industry and sub-vertical, revenue/employee size band, geography/region, business model (B2B/B2C/B2G), regulatory context, buyer roles/titles, use cases, pain points, and deal-size ranges. Route these into the Ideal Customer Profile, Demand Signals, and Buyer Personas sections — extracting this pattern is the whole point of reading the collateral.

Examples:

  • "Acme Regional Bank ($4B assets, Ohio)" -> "mid-market regional US bank, ~$4B in assets, Midwest" (ICP signal)
  • "cut Globex's claims cycle 40%" -> "cut a mid-market insurer's claims cycle ~40%" (anonymized proof point)
  • "Jane Doe, CFO, sponsored the pilot" -> "the CFO sponsored the pilot" (buyer-persona signal)

The offering itself — its name, capabilities, and the seller/vendor — is the SUBJECT of the analysis and is NOT anonymized. Only the clients, customers, prospects, and third parties described in the materials are.

If stripping an identity would leave a claim meaningless, keep the anonymized substance (the metric, outcome, or segment) and drop the attribution — never invent a substitute name. </entity_anonymization>

Read the full file on GitHub · 174 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. 2d ago First seen · 174 lines · 95 tokens per session scan A 290611eec93e

Subscribe to this mod's changes

collateral-analyzer is an agent published in the GitHub repository jasonstrimpel/lead-genius-plugin (10 stars, last pushed 1mo ago), licensed MIT. It adds 95 tokens to every session and 2,042 once invoked, about $0.0005 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.