AGENTS.mdA
AGENTS.md from x1jiang/clawagents_py is an AGENTS.md file: the standing instructions coding agents read in that repo.
mdr add x1jiang/clawagents_py/AGENTS.md@git:20260912.c730120Pins git:20260912.c730120 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 14 of 14 checks passed
- Kind
- AGENTS.md file
- Repository
- x1jiang/clawagents_py
- Path
AGENTS.md- Versions recorded
- 2
- Latest version
- git:20260912.c730120, committed Sep 12, 2026
- First version recorded
- Apr 25, 2026
- Size
- 5,139 bytes
- Repository stars
- 18
- License
- MIT
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# AGENTS.md — clawagents (Python) Operational guide for AI/automation agents working *on* the `clawagents` Python codebase. This file is intentionally short. It captures the conventions that are easy to violate accidentally and hard to fix retroactively. ## 1. Prompt-cache policy Every long-running agent session relies on the LLM provider's **prompt prefix cache**. A handful of slash-commands and tools mutate state that lives in the system prompt (skills, permission mode, persona, model routing, …). If those mutations apply *immediately*, the cached prefix is invalidated and the next turn pays the full re-prefill cost. For multi-turn tool runs this typically dominates latency and cost. The policy mirrors the one we adopted from Hermes: - **`cache_impact = "none"`** — read-only commands. Always safe mid-run. Examples: `/help`, `/status`, `/version`, `/tools`, `/models`, `/trace`, `/profile`, `/redact`, `/history`. - **`cache_impact = "deferred"`** — *default* for any command that mutates system-prompt state. The change is staged for the **next session** so the current prefix cache survives. Pass `--now` to opt into immediate …
Read the whole file at its exact version.
How to install
mdr add x1jiang/clawagents_py/AGENTS.md@sha256:5268bb7a1dae76c3The line at the top of this page pins the label git:20260912.c730120, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_7bzkmyvrn3mlyszq)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, between Apr 25, 2026 and Sep 12, 2026, about one every 140 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260912.c730120 latest | 2026-09-12 | c730120 | 5,139 B | A | view · diff |
| git:20260425.1407251 | 2026-04-25 | 1407251 | 4,715 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (5139 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
x1jiang/clawagents_py · 18 stars · license MIT · pushed 2026-10-29 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_7bzkmyvrn3mlyszq GET https://markdownregistry.com/api/v1/resolve?ref=x1jiang/clawagents_py/AGENTS.md GET https://markdownregistry.com/api/v1/blob/5268bb7a1dae76c308a6cb49c567a06625cace60eb65b5ba8dd29e638731afb1
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: how to write a good AGENTS.md · AGENTS.md vs CLAUDE.md · which agent file does what.
More from x1jiang/clawagents_py
Every file in x1jiang/clawagents_py