resumasher CLAUDE.md

Project instructions for agents working on the resumasher repository. They explain how to handle the project’s Python environments when the agent runs in a sandbox or container.

In plain words
What is it for?
Choosing the correct virtual environment, bootstrapping a sandbox-local one when needed, and following resumasher-specific development guidance.
Why use it?
They help prevent changes to a host machine’s virtual environment from breaking it or from using binaries that cannot run in the agent’s environment.

Instructions file

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/earino/resumasher/claude-md
Clone the repo
git clone --depth 1 https://github.com/earino/resumasher
Per session 1,157 This file is loaded in full into every session.
When invoked 1,157 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.

ModelPer sessionOnce invoked
Fable 5 $0.01157 $0.01157
Opus 5 $0.00579 $0.00579
Sonnet 5 $0.00231 $0.00231
Haiku 4.5 $0.00116 $0.00116

Measured yesterday against content hash 393833b3e2c3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

resumasher 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 yesterday.

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 · 112 lines

How it starts

The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Project guidance for Claude Code agents

This file is loaded automatically when an agent starts work in this repo. It documents conventions specific to working on resumasher as Claude.

Sandbox vs host venv coordination

This repo's .venv/ is a per-platform artifact owned by whoever installed the project on the host (typically macOS or Windows native, via install.sh). It contains platform-specific Python binaries (Linux ELF, macOS Mach-O, or Windows PE) baked-in paths in pyvenv.cfg, etc.

If you are a sandbox / Docker / remote-container agent sharing the host's filesystem: do NOT write to .venv/. Doing so will either silently break the host's installation (you'll overwrite their venv with one whose binaries can't execute on their platform) or your own builds will silently break (you'll inherit a venv whose binaries can't execute in your sandbox).

What to do instead

Use a sandbox-local venv at $HOME/.venv-resumasher-sandbox/. $HOME is typically container-local in Docker (not part of the shared mount), so the two venvs never collide.

If the sandbox venv doesn't exist yet, bootstrap it with:

bash tools/bootstrap-sandbox-venv.sh

This is idempotent (no-op if it exists), takes ~30s on a fresh container.

How tools find the right venv

bin/resumasher-exec searches in this order:

  1. $RESUMASHER_VENV/bin/python — explicit override (CI, custom setups)
  2. $HOME/.venv-resumasher-sandbox/bin/python — sandbox convention
  3. $SKILL_ROOT/.venv/bin/python — POSIX host install
  4. $SKILL_ROOT/.venv/Scripts/python.exe — Windows host install

So once tools/bootstrap-sandbox-venv.sh has run, all wrapper invocations do the right thing automatically. Hosts whose .venv/ works keep using it unchanged.

When to use which

  • Running tests for development iteration in a sandbox: invoke ~/.venv-resumasher-sandbox/bin/python -m pytest -q directly. Faster than going through the wrapper.
  • Anything that involves cross-process invocation of project scripts (subprocess from a test, SKILL.md $RS calls, etc.): always go through bin/resumasher-exec. It picks the right venv per the search order.

Read the full file on GitHub · 112 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.

  1. yesterday First seen · 112 lines · 1,157 tokens per session scan A 393833b3e2c3

Subscribe to this mod's changes

resumasher CLAUDE.md is an instructions file published in the GitHub repository earino/resumasher (7 stars, last pushed 3mo ago), licensed MIT. It adds 1,157 tokens to every session, about $0.0058 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.