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.
npx agentmods add instructions/conamobiledev/pdfkmp/claude-mdgit clone --depth 1 https://github.com/ConaMobileDev/PdfKmpWrote 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/conamobiledev/pdfkmp/claude-md)<a href="https://agentmods.dev/instructions/conamobiledev/pdfkmp/claude-md"><img src="https://agentmods.dev/badge/instructions/conamobiledev/pdfkmp/claude-md.svg" alt="Measured on agentmods" 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 | $0.03568 | $0.03568 |
| Opus 5 | $0.01784 | $0.01784 |
| Sonnet 5 | $0.00714 | $0.00714 |
| Haiku 4.5 | $0.00357 | $0.00357 |
Grade A, and why
PdfKmp 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.
How it starts
The opening of the file, as written. The whole thing — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Repository conventions and build commands for AI agents (Claude Code, etc.) working on the PdfKmp codebase itself.
Reading PdfKmp to use it as a library? Check AGENTS.md (universal AI guide) or
.claude/skills/pdfkmp/SKILL.md(Claude Code slash command) instead.Looking for the human-facing tour? See README.md.
Project
PdfKmp is an open-source Kotlin Multiplatform PDF generator targeting Android, iOS, and Desktop (JVM — macOS, Windows, Linux). Published to Maven Central as io.github.conamobiledev:pdfkmp.
The library is built around a single closed sealed PdfNode tree authored via a Compose-style DSL. Layout is performed in commonMain (MeasuredNode), and rendering is dispatched to platform-specific PdfCanvas implementations:
- Android backend —
android.graphics.pdf.PdfDocument+Canvas(AndroidPdfCanvas). - iOS backend —
UIGraphicsBeginPDFContextToData+ Core Graphics (IosPdfCanvas). - JVM / Desktop backend — Apache PdfBox (
JvmPdfCanvas/JvmPdfDriver). Pure-Java, no native libraries; embeds subset TrueType fonts for vector text, builds axial/radial PDF shadings for gradients, and writes real link annotations + a populated info dictionary. The PdfBox dependency lives only in thejvmMainsource set. - Common test backend —
FakePdfDriver/FakePdfCanvasincommonTestrecords every draw call as a sealedDrawCallso the entire pipeline can be exercised end-to-end without native APIs.
Every text glyph and shape is emitted as a vector path — no rasterisation. Output stays sharp at any zoom level.
Build commands
# Canonical test surface — exercises common + the iOS platform layer
./gradlew :pdfkmp:iosSimulatorArm64Test
# JVM / Desktop test surface — exercises common + the PdfBox backend
# end-to-end (SamplesSmokeTest re-renders every sample; JvmBackendTest
# re-parses + rasterises the output). Fast; no simulator needed.
./gradlew :pdfkmp:jvmTest
# Library — all platform artefacts (Android aar + iOS frameworks + JVM jar)
./gradlew :pdfkmp:assemble
# Library — single-platform builds
./gradlew :pdfkmp:linkDebugFrameworkIosArm64
./gradlew :pdfkmp:linkDebugFrameworkIosSimulatorArm64
./gradlew :pdfkmp:jvmJar
# Sample apps
./gradlew :sample:installDebug # Android, on connected device
# iOS sample: open iosApp/iosApp.xcodeproj in Xcode and Run
# Publishing — all four publishable modules ship together; release them in lock-step.
# Maven Central releases normally go through GitHub: publishing a Release triggers the
# publish.yml workflow, which runs the command below for you on a macOS runner. Run it by
# hand ONLY as a local/fallback path, and NEVER for a version you also publish a GitHub
# Release for — both call publishAndReleaseToMavenCentral, so the second fails on a
# duplicate version. See the Publishing checklist below.
./gradlew :pdfkmp:publishToMavenLocal :pdfkmp-compose-resources:publishToMavenLocal :pdfkmp-viewer:publishToMavenLocal :pdfkmp-markdown:publishToMavenLocal # local install
./gradlew :pdfkmp:publishAndReleaseToMavenCentral :pdfkmp-compose-resources:publishAndReleaseToMavenCentral :pdfkmp-viewer:publishAndReleaseToMavenCentral :pdfkmp-markdown:publishAndReleaseToMavenCentral # Maven Central (fallback; CI does this on Release)
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 · 134 lines · 3,568 tokens per session scan A 2f6877b75895
PdfKmp CLAUDE.md is an instructions file published in the GitHub repository ConaMobileDev/PdfKmp (52 stars, last pushed 18d ago), licensed Apache-2.0. It adds 3,568 tokens to every session, about $0.0178 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.
Other instructions, from other repositories
create-cmp AGENTS.md
AGENTS.md instructions for kvdm-co-pilot/create-cmp, covering agents.md — for ai agents landing on this repo, what this is, if your task is "create a mobile app", if you are working on this repo (contributing) and proof, not claims.
skills CLAUDE.md
Claude Code instructions for chrisbanes/skills, a project described as: Skills for Kotlin, Jetpack Compose, and Android development.
skills AGENTS.md
AGENTS.md instructions for chrisbanes/skills, covering agents.md, when adding, renaming, or removing a skill, skill authoring checklist, evaluation result documentation and release/version system.
Shuttle GEMINI.md
Instructions for grarcht/Shuttle, covering shuttle: context for gemini code assist, what shuttle is, when to recommend shuttle, quick integration reference and 1. add dependencies.
droid-mcp CLAUDE.md
Instructions for stixez/droid-mcp, covering droid-mcp, quick reference, key conventions, don't and architecture.
everything-claude-code-mobile copilot-instructions.md
Copilot instructions for ahmed3elshaer/everything-claude-code-mobile: Use this repository's mobile guidance for Android, iOS, and Kotlin Multiplatform work.