What is selectools CLAUDE.md?

Project instructions for Selectools, a Python library for AI assistants with tool calling, retrieval-augmented generation, and multi-agent coordination. It supports several model providers and includes tests, documentation, and type checking.

Use them when changing Selectools code or documentation, running tests, formatting and linting, checking types, reviewing security, or working on model and agent integrations.

They collect the project’s commands, layout, and known pitfalls in one place. This helps an AI assistant work within the library’s development and quality checks.

selectools CLAUDE.md — Instructions file by johnnichev

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/johnnichev/selectools/claude-md

Clone the repo

git clone --depth 1 https://github.com/johnnichev/selectools

Per session 1,672 This file is loaded in full into every session.

When invoked 1,672 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.

Model	Per session	Once invoked
Fable 5	$0.01672	$0.01672
Opus 5	$0.00836	$0.00836
Sonnet 5	$0.00334	$0.00334
Haiku 4.5	$0.00167	$0.00167

Measured 2d ago against content hash 8dceda451d74, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

selectools 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 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.

CLAUDE.md · 73 lines

How it starts

The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Selectools — Agent Instructions

Production-ready Python library for AI agents with tool calling, RAG, and multi-agent orchestration. OpenAI, Azure OpenAI, Anthropic, Gemini, Ollama, LiteLLM (100+ models), RouterProvider. Python 3.9+, src-layout, pytest, MkDocs Material.

Version: src/selectools/__init__.py
PyPI: https://pypi.org/project/selectools/
Docs: https://selectools.dev

See AGENTS.md for commands, boundaries, and condensed landmines. See subdirectory CLAUDE.md files for scoped rules: tests/, src/selectools/, docs/.

Session workflow: At ~40 messages or when context feels heavy: update HANDOFF.md with current state, run /clear, start fresh. This outperforms auto-compaction.

Commands

pytest tests/ -x -q                                    # All tests
pytest tests/ -k "not e2e" -x -q                       # Skip E2E
ruff format src/ tests/                                 # Format (replaces Black + isort)
ruff check src/ tests/ --fix                            # Lint + auto-fix (replaces flake8)
mypy src/                                               # Type check
bandit -r src/ -ll -q -c pyproject.toml                 # Security
cp CHANGELOG.md docs/CHANGELOG.md && mkdocs build       # Docs

Stability Markers

Apply to every public class/function in __init__.py exports:

Marker	When
`@stable`	Core APIs, at least one release proven
`@beta`	First release, experimental
`@deprecated(since="0.X", replacement="Y")`	Removing/renaming. Keep 2 minor versions

StepType Reference (27 types)

llm_call, tool_selection, tool_execution, cache_hit, error, structured_retry, guardrail, coherence_check, output_screening, session_load, session_save, memory_summarize, entity_extraction, kg_extraction, budget_exceeded, cancelled, prompt_compressed, graph_node_start, graph_node_end, graph_routing, graph_checkpoint, graph_interrupt, graph_resume, graph_parallel_start, graph_parallel_end, graph_stall, graph_loop_detected

Read the full file on GitHub · 73 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.

2d ago First seen · 73 lines · 1,672 tokens per session scan A 8dceda451d74

Subscribe to this mod's changes

selectools CLAUDE.md is an instructions file published in the GitHub repository johnnichev/selectools (11 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 1,672 tokens to every session, about $0.0084 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.

AI Safety Debugging Guardrails Security Testing ai-agents anthropic enterprise-ai gemini

selectools CLAUDE.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Selectools — Agent Instructions

Commands

Stability Markers

StepType Reference (27 types)

What this file has done since we first saw it

Other instructions, from other repositories

Doberman-Core AGENTS.md

agent-guardrails-template copilot-instructions.md

agent-guardrails-template CLAUDE.md

FeatherlaneAI AGENTS.md

agent47 AGENTS.md

agent47 CLAUDE.md