Borrowing it
Nothing to install: this file belongs to FranRom/pupila. 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/FranRom/pupila/main/CLAUDE.mdgit clone --depth 1 https://github.com/FranRom/pupilaWrote 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/franrom/pupila/claude-md)<a href="https://agentmods.dev/instructions/franrom/pupila/claude-md"><img src="https://agentmods.dev/badge/instructions/franrom/pupila/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/franrom/pupila/claude-md"><img src="https://agentmods.dev/badge/instructions/franrom/pupila/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 | $0.04108 | $0.04108 |
| Opus 5.5 | $0.01643 | $0.01643 |
| Sonnet 5.5 | $0.00822 | $0.00822 |
| Haiku 4.5 | $0.00411 | $0.00411 |
Grade A, and why
pupila 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 25d 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 — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for future Claude Code sessions working in this repo. Slim by design — this file is loaded every turn. Subsystem details live next to the code that needs them; procedural how-tos live in skills.
Where things live
| Topic | Source |
|---|---|
| Adding a fetcher / tier-S slug | pupila-fetchers skill |
Tuning filter weights / debugging _signals |
pupila-filters skill |
| AI review + AI Apply pipelines | pupila-ai-review skill |
UI patterns (CSS modules, lib/api/, hooks, perf) |
ui/CLAUDE.md (auto-loaded in ui/) |
| MCP server invariants + add-a-tool recipe | src/mcp/CLAUDE.md (auto-loaded in src/mcp/) |
| Fetcher security checklist | src/fetchers/CLAUDE.md (auto-loaded in src/fetchers/) |
Overview
pupila is a config-driven, forkable daily job aggregator. Fetches from 19 public sources (5 ATS APIs — Ashby, Greenhouse, Lever, Recruitee, Personio (XML) — plus RSS, JSON boards, Hacker News, HTML scrapers, an Aave Next.js scraper, ashby-private for orgs whose public posting-API is disabled, jobicy and himalayas — no-key remote-jobs feeds with region/country-tagged locations and structured salary, and bluedoor — a free aggregator over ~1.6M postings/31 ATS, queried by profile.location), normalizes them, applies hard exclusion filters, computes a per-job fitScore, deduplicates, and writes data/jobs.json, an RSS feed at data/feed.xml, and an auto-regenerated JOBS.md. README.md is hand-maintained — never overwrite it from code. No external services, no DB.
Cross-cutting invariants (apply repo-wide):
config/profile.jsonis gitignored — encodes personal scoring preferences. Auto-bootstraps from committedconfig/profile.default.jsonon firstpnpm run dev/pnpm run uiviabootstrapProfileIfMissing()(idempotent —COPYFILE_EXCLno-ops on the steady state). Don't bypass; don't commit personalized weights.- Mandatory CV gate:
pnpm run devchecks forconfig/candidate-brief.mdat startup and exits 1 if missing. Bypass withPUPILA_NO_BRIEF_CHECK=1or--no-brief-check. config/candidate-brief.mdis the only natural-language config (gitignored). Generated viapnpm run setup-brief --file ~/cv.pdfor via the UI's Profile tab (drop PDF/DOCX/MD CV). CLI shells out toclaude/codex/gemini/opencode/ollama— auto-detected, overridePUPILA_LLM=<provider>(ollama model via UI selection /PUPILA_LLM_MODEL, or the sole pulled model with a warning; no hardcoded default).- Local-first scheduling: daily aggregation runs via
scripts/install-launchd.sh(macOS) orscripts/install-cron.sh(Linux), not GitHub Actions cron. CI runs only on push/PR for gates. config/slugs.local.jsonis gitignored: a personal overlay of per-ATS company-slug add/removes layered on the committedconfig/slugs.jsonbaseline (effective list = shipped baseline plus overlay adds, minus overlay removes; resolved at fetch time insrc/lib/slugs.ts). Written only by the UI's Settings → Job sources panel (/api/sources); never writeconfig/slugs.jsonfrom app code. See thepupila-fetchersskill.config/applied.jsonsource of truth for application tracking (gitignored personal data; the UI writes it via Vite middleware). It's your private application history, so it's NOT committed (that would leak it to anyone cloning the repo); copy the file if you need it on another machine. Don't filter applied jobs out of the main list: the user wants them visible.
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.
- 25d ago Changed · +68 tokens per session a7e3158886b0
- 1mo ago First seen · 202 lines · 4,040 tokens per session scan A a23386431724
pupila CLAUDE.md is an instructions file published in the GitHub repository FranRom/pupila (36 stars, last pushed yesterday), licensed MIT. It adds 4,108 tokens to every session, about $0.0164 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 2026-08-30.
Other instructions, from other repositories
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).
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.