What is pbi-context CLAUDE.md?

Project instructions for pbi-context, a tool for working with Power BI project files. PBIP and TMDL are text-based file formats used to store parts of Power BI reports and models.

Use them when planning or changing pbi-context; they require a plan and explicit approval before code is modified.

They give coding agents the project's current requirements and history, including required PBIP/TMDL support, so agents do not repeat earlier research or skip established constraints.

pbi-context CLAUDE.md — Instructions file for coding agents by Osc2405: Project instructions for pbi-context, a tool for working with…

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/osc2405/pbi-context/claude-md

Clone the repo

git clone --depth 1 https://github.com/Osc2405/pbi-context

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/osc2405/pbi-context/claude-md.svg)](https://agentmods.dev/instructions/osc2405/pbi-context/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/osc2405/pbi-context/claude-md"><img src="https://agentmods.dev/badge/instructions/osc2405/pbi-context/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 16,088 This file is loaded in full into every session.

When invoked 16,088 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5	$0.16088	$0.16088
Opus 5	$0.08044	$0.08044
Sonnet 5	$0.03218	$0.03218
Haiku 4.5	$0.01609	$0.01609

Measured 4d ago against content hash 77d96ac0692a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pbi-context CLAUDE.md 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 4d 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.

CLAUDE.md · 858 lines

How it starts

The opening of the file, as written. The whole thing — 858 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — pbi-context: soporte PBIP/TMDL + mejoras de contexto para agentes

Renombrado de pbi-docs a pbi-context el 2026-08-03 — ver la entrada fechada al final de este archivo para el porqué. Las entradas fechadas anteriores a esa se dejan tal cual (narran hechos de cuando el proyecto se llamaba pbi-docs), consistente con el criterio de no reescribir historia que este mismo archivo ya aplica en otras partes.

Este archivo es la fuente de verdad para las próximas fases de desarrollo de pbi-context. Léelo completo antes de proponer un plan. Trabajar siempre en Plan Mode antes de tocar código: presentar el plan, esperar aprobación explícita, y solo entonces implementar por fases.

Estado actual (actualizado 2026-07-15)

Las secciones 1 y 2 de este documento describían trabajo planeado que ya está implementado y probado (161 tests, 0 dependencias externas). Se dejan las secciones originales abajo sin reescribir — el razonamiento de diseño sigue siendo válido y útil — pero marcadas como completadas para que una sesión futura no las re-planifique desde cero.

Sección 1 (soporte PBIP/TMDL): ✅ COMPLETADO. pbi_extractor/pbip_extractor.py. Validado contra un export .pbip real (no solo el fixture sintético) — ver docs/pbip_validation_report.md para los 5 bugs encontrados y corregidos en el proceso.
Sección 2 (salida indexada + TOON): ✅ COMPLETADO. pbi_extractor/indexed_output.py, pbi_extractor/toon_encoder.py. Medido empíricamente, no solo implementado — ver docs/token_optimization_report.md (TOON no es un ahorro uniforme, gana solo en tablas grandes/uniformes; el ahorro dominante viene de index.json + carga selectiva, no del formato).
Trabajo adicional no anticipado en este documento original, construido sobre lo anterior: pbi_extractor/resolver.py (capa de consulta estructurada, JSON/TOON transparente), servidor MCP de solo lectura hecho a mano (pbi_extractor/mcp_server.py, sin dependencia del SDK mcp oficial para preservar cero-dependencias), Skills invocables desde Claude Code y GitHub Copilot (.claude/skills/analyze-pbi-model/, .github/prompts/analyze-pbi-model.prompt.md), y un experimento de precisión automatizado (docs/precision_validation_report.md) que valida que el contexto acotado no sacrifica exactitud de respuesta.
Documento de planeación estratégica activo: docs/Analisis_Posicionamiento_Comparativa_Plan_Futuro.md — reemplaza al checklist original de esta sesión de Plan Mode como fuente de verdad del roadmap (secciones 3-4 de ese documento). Secciones 3 y 4 de este archivo (Graphify, Grafo diferido) siguen vigentes sin cambios.

Read the full file on GitHub · 858 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.

4d ago First seen · 858 lines · 16,088 tokens per session scan A 77d96ac0692a

Subscribe to this mod's changes

pbi-context CLAUDE.md is an instructions file published in the GitHub repository Osc2405/pbi-context (15 stars, last pushed 1mo ago), licensed MIT. It adds 16,088 tokens to every session, about $0.0804 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 2026-08-30.

Analytics and metrics LLM ai-agents dax documentation-generator mcp mcp-server

pbi-context CLAUDE.md

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

CLAUDE.md — pbi-context: soporte PBIP/TMDL + mejoras de contexto para agentes

Estado actual (actualizado 2026-07-15)

What this file has done since we first saw it

Other instructions, from other repositories

blockrun-mcp AGENTS.md

GhidrAssistMCP CLAUDE.md

openrouter-mcp-multimodal AGENTS.md

intervals-icu-mcp CLAUDE.md

pfsense-mcp-server AGENTS.md

dremio-mcp AGENTS.md