salesagent CLAUDE.md

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 instructions/prebid/salesagent/claude-md

Clone the repo

git clone --depth 1 https://github.com/prebid/salesagent

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/prebid/salesagent/claude-md.svg)](https://agentmods.dev/instructions/prebid/salesagent/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/prebid/salesagent/claude-md"><img src="https://agentmods.dev/badge/instructions/prebid/salesagent/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 8,568 This file is loaded in full into every session.

When invoked 8,568 The same file — it is already loaded in full.

Security scan A 1 finding. 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.

Model	Per session	Once invoked
Fable 5	$0.08568	$0.08568
Opus 5	$0.04284	$0.04284
Sonnet 5	$0.01714	$0.01714
Haiku 4.5	$0.00857	$0.00857

Measured 2d ago against content hash 9d571ebf5916, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

salesagent CLAUDE.md 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.

fetch(apiUrl, { credentials: 'same-origin' });

CLAUDE.md · 767 lines

How it starts

The opening of the file, as written. The whole thing — 767 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prebid Sales Agent - Development Guide

🤖 For Claude (AI Assistant)

This guide helps you work effectively with the Prebid Sales Agent codebase maintained under Prebid.org. Key principles:

Working with This Codebase

Always read before writing - Use Read/Glob to understand existing patterns
Test your changes - Run make quality before committing
Follow the patterns - 7 critical patterns below are non-negotiable
When stuck - Check /docs for detailed explanations
Pre-commit hooks are your friend - They catch most issues automatically
Name your PRs correctly - they need to pass .github/workflows/pr-title-check.yml

Common Task Patterns

Adding a new AdCP tool: Extend library schema → Add _impl() function → Add MCP wrapper → Add A2A raw function → Add tests
Fixing a route issue: Check for conflicts with grep -r "@.*route.*your/path" → Use url_for() in Python, scriptRoot in JavaScript
Modifying schemas: Verify against AdCP spec → Update Pydantic model → Run pytest tests/unit/test_adcp_contract.py
Database changes: Use SQLAlchemy 2.0 select() → Use JSONType for JSON → Create migration with alembic revision

Key Files to Know

src/core/main.py - MCP server and tool registration
src/core/tools/ - tool _impl() business logic, MCP wrappers, and A2A raw functions (package)
src/core/schemas/ - Pydantic models, AdCP-compliant (package)
src/adapters/base.py - Adapter interface
src/adapters/gam/ - GAM implementation
tests/unit/test_adcp_contract.py - Schema compliance tests

DRY (Don't Repeat Yourself) — Non-Negotiable Invariant

DRY is not "premature optimization." It is not "refactoring beyond what was asked." It is a correctness requirement, equivalent to type safety or test integrity.

If you write a block of logic that is structurally similar to an existing block (same pattern, different variables), you MUST extract a shared helper function
If you are asked to refactor duplicated code, that is a bug fix, not an "improvement"
NEVER cite "avoid over-engineering" or "keep it simple" to justify leaving duplicated logic in place
Duplicated code is a defect. It means the next person who fixes a bug in one copy will miss the other copy. This is not theoretical — it has caused real bugs in this codebase
Enforced by: check_code_duplication.py pre-commit hook (pylint R0801, ratcheting baseline in .duplication-baseline)

Read the full file on GitHub · 767 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.

2d ago First seen · 767 lines · 8,568 tokens per session scan A 9d571ebf5916

Subscribe to this mod's changes

salesagent CLAUDE.md is an instructions file published in the GitHub repository prebid/salesagent (36 stars, last pushed 2d ago), licensed Apache-2.0. It adds 8,568 tokens to every session, about $0.0428 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 2026-09-01.

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens