Borrowing it
Nothing to install: this file belongs to SFHAJJI/energy-copilot. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/SFHAJJI/energy-copilot/main/CLAUDE.mdgit clone --depth 1 https://github.com/SFHAJJI/energy-copilotWrote 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/instructions/sfhajji/energy-copilot/claude-md)<a href="https://agentmods.dev/instructions/sfhajji/energy-copilot/claude-md"><img src="https://agentmods.dev/badge/instructions/sfhajji/energy-copilot/claude-md/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/instructions/sfhajji/energy-copilot/claude-md"><img src="https://agentmods.dev/badge/instructions/sfhajji/energy-copilot/claude-md.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 | $0.21702 | $0.21702 |
| Opus 5 | $0.10851 | $0.10851 |
| Sonnet 5 | $0.04340 | $0.04340 |
| Haiku 4.5 | $0.02170 | $0.02170 |
Grade A, and why
energy-copilot 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 9d 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 — 615 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Luxembourg Energy Prospection Co-Pilot — Technical Specification v2.1 (build)
Owner: Soufien · Date: 19 Jul 2026 · Status: build-ready · Companion docs: copilot-datasets-definitive.md, blockers-todo.md, copilot-functional-specs.md (v1, superseded where this conflicts)
0. Scope deltas vs v1
- Sharing Simulator: DROPPED. Remove from build, site, and evals. (Leneda registry dataset stays — it feeds the whitespace layer; the Leneda metering API mock goes.)
- Rules Expert = Creos technical rules only (TABs + notes d'information + communal/national aids). No laws.
- NEW — Law Expert: all Luxembourg law (Legilux) + EU law (EUR-Lex/CELLAR), built as a standalone reusable service (
lu-law-service): own repo, own index, exposed as MCP server + REST. This co-pilot mounts it as a tool; the future finance/trading co-pilot mounts the same service with different theme scopes. - Tender Watcher is in-build (guichet channel first; TED second; pmp newsletter third).
- Experts in v2: Orchestrator + Roof Hunter (finder family) + Rules Expert + Law Expert + Watcher (pipeline).
1. UI / UX
Pattern: map-first split workspace (real-estate-style list⟷map sync), one chat, stage-based navigation.
- Navigation = workflow stages, not specialists:
Prospection·Développement·Veille·Engine Room. A chip on every answer still shows answered by: . - Prospection (home): full-height MapLibre GL map of Luxembourg (CARTO basemap) + right-hand card list, fully synchronized: hover card ↔ highlight pin, click → detail drawer, map pan/zoom refilters the list. Layer toggles: roofs · parkings (ombrières) · HV backbone (WMS) · whitespace choropleth (commune) · [agri mode: LIS-L land use, PAG zones, Natura/biotope exclusions] · greyed-out "Distribution network — requires Creos data".
- Candidate card fields: ortho thumbnail, address, commune, usable m² / kWc / MWh / €·yr (each with a lineage chip: source + confidence
verified | computed | to-verify | closed), suitability class,pv_visibleflag (orthophoto-vision, photo vintage shown — §19 D10), occupant candidate (OSM, "to verify") with links: Editus company page + LBR/RCS (links, never scraped — no open contact dataset exists, see §4 note), rule flags from Rules Expert, buttons:→ Dossier·→ ask Rules·→ ask Law. - Context ribbon: when a call is selected in Veille, a persistent ribbon shows its parsed constraints (kWc band, deadline, lots). Every finder and every expert inherits the ribbon's constraint object — the visible form of shared session memory. One click clears it.
- Développement: the Dossier view — a structured feasibility study per candidate: pre-filled public fields (address/id_caclr, kWc/MWh/€, PAG zoning, matched call + aid stack incl. commune top-up, connection flags), an orthophoto pane for the human roof-covering pre-screen, and labeled human-input fields (matériau, descente de charges, amiante risk, étanchéité, occupant consumption, Creos quote). Business case recalculates as fields land. Export: PDF.
- Veille: calls table (family: aide | marché; status; deadline; source links incl. cahier/FAQ/results), subscription manager (NL constraint → parsed → email, double opt-in), per-call page with
find candidates for this call. - Engine Room: Quality (eval scores per expert + history), Status, AI Act & Privacy, Changelog, About (architecture + repos).
- Chat: persistent bottom bar on every stage; each answer: citations, answered-by chip,
how was this made?trace viewer, 👍/👎. - Languages: FR/EN UI; German citations rendered as-is with the note "version allemande juridiquement déterminante (TAB)".
- Mobile: map collapses behind a toggle; cards stack.
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.
- 9d ago First seen · 615 lines · 21,702 tokens per session scan A 75e0c163547c
energy-copilot CLAUDE.md is an instructions file published in the GitHub repository SFHAJJI/energy-copilot (2 stars, last pushed 28d ago), licensed Apache-2.0. It adds 21,702 tokens to every session, about $0.1085 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 instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.