AGENTS.md@benchmark/MR-NIAHA
AGENTS.md@benchmark/MR-NIAH from unvulcanised-watercress762/mem9 is an AGENTS.md file: the standing instructions coding agents read in that repo.
mdr add unvulcanised-watercress762/mem9/benchmark/MR-NIAH/AGENTS.md@git:20260312.12684e6Pins git:20260312.12684e6 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 14 of 14 checks passed
- Kind
- AGENTS.md file
- Repository
- unvulcanised-watercress762/mem9
- Path
benchmark/MR-NIAH/AGENTS.md- Versions recorded
- 3
- Latest version
- git:20260312.12684e6, committed Mar 12, 2026
- First version recorded
- Mar 10, 2026
- Size
- 2,412 bytes
- Repository stars
- 3
- License
- Apache-2.0
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
## Overview MR-NIAH is a bridge from the MiniMax benchmark corpus to OpenClaw sessions and mem9 comparison runs. ## Files and workflow | File | Role | |------|------| | `fetch_data.py` | Mirror/update upstream dataset into `origin/` | | `mr-niah-transcript.py` | Convert raw turns into OpenClaw session JSON | | `run_batch.py` | Replay generated sessions through an OpenClaw profile | | `run_mem_compare.sh` | Compare baseline vs mem9-enabled profile | | `score.py` | Apply MR-NIAH scoring rubric to predictions | | `USAGE.md` | Full prerequisites and end-to-end usage | ## Where to look - Dataset cache and raw source: `origin/` - Generated sessions and index: `output/` - Latest run outputs: `results/` - Preserved comparison outputs: `results-*/` - Helper state: `.cache/` ## Commands ```bash cd benchmark/MR-NIAH && python3 fetch_data.py cd benchmark/MR-NIAH && python3 mr-niah-transcript.py cd benchmark/MR-NIAH && python3 run_batch.py --profile mrniah_local --agent main --local --limit 30 cd benchmark/MR-NIAH && MRNIAH_LIMIT=30 bash run_mem_compare.sh cd benchmark/MR-NIAH && python3 score.py results/predictions.jsonl ``` ## Local conventions …
Read the whole file at its exact version.
How to install
mdr add unvulcanised-watercress762/mem9/benchmark/MR-NIAH/AGENTS.md@sha256:8e5e659b1b0f03acThe line at the top of this page pins the label git:20260312.12684e6, 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_gmt6w6zs4upop6eb)
5 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Mar 10, 2026 and Mar 12, 2026, about one every day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260312.12684e6 latest | 2026-03-12 | 12684e6 | 2,412 B | A | view · diff |
| git:20260311.e617ba8 | 2026-03-11 | e617ba8 | 2,340 B | A | view · diff |
| git:20260310.1bcdba2 | 2026-03-10 | 1bcdba2 | 1,467 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (2412 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
unvulcanised-watercress762/mem9 · 3 stars · license Apache-2.0 · pushed 2026-10-14 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_gmt6w6zs4upop6eb GET https://markdownregistry.com/api/v1/resolve?ref=unvulcanised-watercress762/mem9/benchmark/MR-NIAH/AGENTS.md GET https://markdownregistry.com/api/v1/blob/8e5e659b1b0f03ac0e9dec48f6dd03408608ea59562e065d775a71e6aece7ea1
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.