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/tlaloc-es/killpy/agents-mdgit clone --depth 1 https://github.com/Tlaloc-Es/killpyWhat 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 | $0.01818 | $0.01818 |
| Opus 5 | $0.00909 | $0.00909 |
| Sonnet 5 | $0.00364 | $0.00364 |
| Haiku 4.5 | $0.00182 | $0.00182 |
Grade A, and why
killpy AGENTS.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 2d 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project Overview
killpy is a Python CLI/TUI utility to find and remove Python-related environments and free disk space.
Main capabilities:
- Discover local virtual environments (
.venvand folders containingpyvenv.cfg) - Discover Conda environments
- Discover Poetry virtualenvs
- Discover installed
pipxpackages and uninstall them - Remove
__pycache__directories
Primary UX:
- Interactive terminal UI (Textual app) for listing and deleting environments
- CLI command to clean caches (
killpy clean)
Goal
Help users reclaim disk space safely by surfacing Python environments with size info and allowing explicit deletion actions.
Tech Stack
- Python
>=3.10(CI/classifiers cover 3.10–3.13; ruff targetspy310) clickfor CLI command routingtextualfor interactive terminal UIrichas a dependency used by Textual ecosystem and terminal rendering
Reference links:
- Textual docs: https://textual.textualize.io/
- Textual repository: https://github.com/Textualize/textual
- Rich docs: https://rich.readthedocs.io/en/latest/introduction.html
- Rich repository: https://github.com/Textualize/rich
Packaging and tooling:
pyproject.toml(PEP 621 metadata)uvsupport ([tool.uv] package = true)- Commit/version workflow via
commitizen - Lint/type/test dev tooling declared under
[dependency-groups].dev
Repository Structure
Top-level:
pyproject.toml— package metadata, dependencies, scripts, tool configREADME.md— user documentationCHANGELOG.md— release history.github/workflows/— CI/release automation
Package layout (killpy/):
__main__.py— CLI entrypoint (killpyscript); thecliclick group, the--delete-allfast path, and subcommand registrationcli.py— Textual app (TableApp) and interactive actionsmodels.py— shared dataclasses:Environment(core contract), plusGitInfo,ScoredEnvironment,Suggestion,ScanRecordscanner.py—Scanner: orchestrates detectors, deduplicates resultscleaner.py—Cleaner: deletion viashutil.rmtree,conda env remove,pipx uninstall, oruv tool uninstall(routed byEnvironment.managed_by)commands/clean.py—killpy cleansubcommandcommands/delete.py—killpy deletesubcommandcommands/list.py—killpy listsubcommandcommands/stats.py—killpy statssubcommandcommands/doctor.py—killpy doctorsubcommand (health report + scoring)commands/find.py—killpy findsubcommand (locate a package across envs)commands/_utils.py— shared command helpers (filter_envs,partition_in_use)intelligence/— scoring layer:git_analyzer.py,scoring.py,suggestions.py,tracker.py(scan-history persistence)cleaners/__init__.py—remove_pycachehelperfiles/__init__.py— file-size helpers (get_total_size,format_size)detectors/— one detector class per environment type:base.py—AbstractDetectorinterfacevenv.py—.venvdirs andpyvenv.cfg-based envspyenv.py— pyenv-managed Python versionspoetry.py— Poetry global virtualenv cacheconda.py— Conda environmentspipx.py— pipx packageshatch.py— Hatch environmentspipenv.py— Pipenv virtualenvstox.py— tox.toxdirectoriesuv.py— uv tool environments (uv tool install) and uv-managed Python installsartifacts.py— build artifacts (dist/,build/,*.egg-info)cache.py— cache dirs (__pycache__,.mypy_cache,.pytest_cache, …)
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.
- 2d ago First seen · 168 lines · 1,818 tokens per session scan A 9f6fcb4f36cd
killpy AGENTS.md is an instructions file published in the GitHub repository Tlaloc-Es/killpy (128 stars, last pushed 14d ago), licensed MIT. It adds 1,818 tokens to every session, about $0.0091 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
ui-ux-pro-max-skill CLAUDE.md
Claude Code instructions for nextlevelbuilder/ui-ux-pro-max-skill, covering claude.md, project overview, search command, architecture and sync rules.
tweakcc AGENTS.md
AGENTS.md instructions for Piebald-AI/tweakcc, covering agents.md, development commands, build, testing and linting & formatting.
mcpc CLAUDE.md
Claude Code instructions for apify/mcpc, covering claude.md, project overview, build and development commands, install dependencies and build the project.
NeoHaskell AGENTS.md
Instructions for neohaskell/NeoHaskell, covering neohaskell — agent guide, hard rule, style (mandatory), commands (pointers, not a duplicate) and work intake.
pinkbin CLAUDE.md
Instructions for cccyd2003-qwq/pinkbin, covering pinkbin · 给 claude 的工作约定, 项目背景, hard rules(不可违反), discovery loop(动手前的纪律) and 实测回环(phase 7 经验).
tokf CLAUDE.md
Instructions for mpecan/tokf, covering tokf — development guidelines, project philosophy, commits, code quality and testing.