AGENTS.md@evalsA

AGENTS.md@evals from max-sixty/tend is an AGENTS.md file: the standing instructions coding agents read in that repo.

Install it, pinned
mdr add max-sixty/tend/evals/AGENTS.md@git:20261010.5ca6cb8

Pins git:20261010.5ca6cb8 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 14 of 14 checks passed
Kind
AGENTS.md file
Repository
max-sixty/tend
Path
evals/AGENTS.md
Versions recorded
9
Latest version
git:20261010.5ca6cb8, committed Oct 10, 2026
First version recorded
Oct 2, 2026
Size
8,546 bytes
Repository stars
39
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

# Promptfoo evals

Codex is the default executor and judge. Default model settings live in `prepare.py`;
`codex-provider.cjs` sets the reasoning effort. Every attempt starts fresh,
without original actor history. Run from the repo root with Node 22.22+ and a
local Codex subscription login:

```bash
npm --prefix evals ci --ignore-scripts
uv run python evals/prepare.py
npm --prefix evals run eval -- --output ../.tmp/evals/results.json
```

The suite compares historical and current plugin guidance, three attempts per
arm, with caching disabled. Both arms use the same current shared system prompt
and model. Preparation replaces staged inputs; do not run it during an eval.
Append `--filter-pattern "[Gg]allery"` to narrow the suite or
`--filter-providers current` to run only current guidance. Failed assertions
make Promptfoo exit nonzero; inspect individual results to distinguish them
from execution errors.

Promptfoo's table has a column per arm (`codex/historical`, `codex/current`)
and a row per case, and every case runs on both arms. Each test's metadata
names its case and kind; the provider reads that case's staged inputs from its
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add max-sixty/tend/evals/AGENTS.md@sha256:b097d36c98ff6362

The line at the top of this page pins the label git:20261010.5ca6cb8, 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_cwshkfvvq7zjs34l.svg)](https://markdownregistry.com/a/art_cwshkfvvq7zjs34l)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 9 versions of this file, between Oct 2, 2026 and Oct 10, 2026, about one every day.

versioncommittedcommitsizeaudit
git:20261010.5ca6cb8 latest2026-10-10 5ca6cb8 8,546 BA view · diff
git:20261010.405d83c2026-10-10 405d83c 7,469 BA view · diff
git:20261006.834e48b2026-10-06 834e48b 5,701 BA view · diff
git:20261005.0c4df1f2026-10-05 0c4df1f 5,971 BA view · diff
git:20261004.166792b2026-10-04 166792b 5,546 BA view · diff
git:20261003.b23fc762026-10-03 b23fc76 5,344 BA view · diff
git:20261002.6fceb2e2026-10-02 6fceb2e 5,744 BA view · diff
git:20261002.a1cd5d02026-10-02 a1cd5d0 4,802 BA view · diff
git:20261002.95427f92026-10-02 95427f9 3,347 BA view

Audit of the latest version

A  14 of 14 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Size between 200 bytes and 200 KB (8546 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

GitHub

max-sixty/tend · 39 stars · license MIT · pushed 2026-10-19 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_cwshkfvvq7zjs34l
GET https://markdownregistry.com/api/v1/resolve?ref=max-sixty/tend/evals/AGENTS.md
GET https://markdownregistry.com/api/v1/blob/b097d36c98ff6362539f71c8381a1a65a9cf3b372ea8148fdb4ba1b1c8422da2

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 max-sixty/tend

release SKILL.md skill
max-sixty/tend · .claude/skills/release/SKILL.md · Tend release workflow. Use when user asks to "do a release", "release a new version", "cut a release", or wants to…
git:20261004.e9e9d01 · audit A · 39 stars
review-reviewers SKILL.md skill
max-sixty/tend · .claude/skills/review-reviewers/SKILL.md · Outcome-based analysis of tend's CI behavior — checks whether tend's outputs were accepted or rejected, escalating to…
git:20261019.13856fe · audit A · 39 stars
running-tend SKILL.md skill
max-sixty/tend · .claude/skills/running-tend/SKILL.md · Tend-specific instructions for tend CI workflows. Adds non-standard workflow inclusion for usage analysis and repo…
git:20261004.92bf109 · audit A · 39 stars
AGENTS.md AGENTS.md file
max-sixty/tend · AGENTS.md
git:20261019.13856fe · audit A · 39 stars
running-tend SKILL.md skill
max-sixty/tend · evals/cases/yolo-merge-authority/policies/leaf/.claude/skills/running-tend/SKILL.md · Project-specific instructions loaded by tend workflows alongside AGENTS.md.
git:20261010.8e7e6dd · audit A · 39 stars
debug-tend-run SKILL.md skill
max-sixty/tend · plugins/install-tend/skills/debug-tend-run/SKILL.md · Debugs a tend CI run from its session-log artifacts. Use for run IDs, PR numbers, or questions about what tend did and…
git:20260929.087f5f1 · audit A · 39 stars
install-tend SKILL.md skill
max-sixty/tend · plugins/install-tend/skills/install-tend/SKILL.md · Installs tend, rotates credentials, repairs failing tend checks, and manages config, secrets, branch protection, and…
git:20261019.13856fe · audit B · 39 stars
act-in-other-repos SKILL.md skill
max-sixty/tend · plugins/tend-ci-runner/skills/act-in-other-repos/SKILL.md · Act in another repository. Use before filing an issue, commenting, or contributing anywhere other than this repo.
git:20260924.b3c4ba6 · audit A · 39 stars
check-requester-access SKILL.md skill
max-sixty/tend · plugins/tend-ci-runner/skills/check-requester-access/SKILL.md · Check a requester's access tier. Use when a request aims you at someone else's work — close, lock, label, revert…
git:20261010.8e7e6dd · audit A · 39 stars
ci-fix SKILL.md skill
max-sixty/tend · plugins/tend-ci-runner/skills/ci-fix/SKILL.md · Debug failing CI on the default branch. Use when CI fails on main.
git:20261019.13856fe · audit A · 39 stars
code-review SKILL.md skill
max-sixty/tend · plugins/tend-ci-runner/skills/code-review/SKILL.md · Structured second pass over a diff — correctness and cleanup angles, a verify pass, ranked findings returned to the…
git:20261004.166792b · audit A · 39 stars
continue-pr SKILL.md skill
max-sixty/tend · plugins/tend-ci-runner/skills/continue-pr/SKILL.md · Continue an unfinished PR with no human author, including red CI and unresolved findings after a prior response or…
git:20261012.364f696 · audit A · 39 stars

Every file in max-sixty/tend

Other files named AGENTS.md@evals

AGENTS.md@evals AGENTS.md file
tencentcloudbase/cloudbase-ai-toolkit · evals/AGENTS.md
git:20260928.b71829c · audit A · 1,141 stars
AGENTS.md@evals AGENTS.md file
theonly1me/shadowclone · evals/AGENTS.md
git:20261010.e7fb7ac · audit A · 21 stars
AGENTS.md@evals AGENTS.md file
b-mendoza/agent-skills · evals/AGENTS.md
git:20260804.71e34db · audit A · 3 stars

Browse by kind, by grade A, or by owner.