iterate-agent-harness SKILL.md skillB

iterate-agent-harness from kensaurus/cursor-kenji is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add kensaurus/cursor-kenji/iterate-agent-harness@git:20260923.68c095e

Pins git:20260923.68c095e in mdr.lock. Pin the fingerprint instead.

Audit grade
B, 16 of 17 checks passed
Kind
SKILL.md skill
Repository
kensaurus/cursor-kenji
Path
skills/iterate-agent-harness/SKILL.md
Versions recorded
5
Latest version
git:20260923.68c095e, committed Sep 23, 2026
First version recorded
Aug 17, 2026
Size
7,128 bytes
Repository stars
9
License
MIT

Turn an agent failure (premature stop, false completion, gamed check, missed file) into a rule, hook, or CI guard with a regression test. Use when "the agent stopped early again", "it gamed the test", or "add a guard so this doesn't recur".

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

# iterate-agent-harness — Failure → Durable Guard

**Degree of freedom: MIXED.** Failure-class and smallest fix `[HIGH freedom]`;
guard-fails-before-fix and the report shape `[LOW freedom — run exactly]`.

When an agent run ends wrong, the fix is not just to redo the task — it is to
change the harness so the whole class of failure cannot recur silently. This
skill converts a single failure into a durable guard plus a regression check.

> **A failure you cannot reproduce, you cannot prove you fixed.** Capture the
> failure mode, add a guard that fails on it, then make the guard pass with the
> smallest durable change.

## How to reason

1. **Observe** — happened vs should-have; transcript, state file, diff, CI
2. **Interpret** — routing miss, missing guard, or a gate that passed wrongly
3. **Classify** — premature stop / false completion / reward hacking / scope / handoff / context loss
4. **Guard** — add a guard that fails on the unfixed behavior, then the smallest durable fix

## Worked example

> **Observe:** agent said complete-everything was done; `.cursor/complete-everything-state.md` still has 12 unchecked items.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add kensaurus/cursor-kenji/iterate-agent-harness@sha256:6dfab1a7b26c0126

The line at the top of this page pins the label git:20260923.68c095e, 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_pyc2jjxzpfrb3ave.svg)](https://markdownregistry.com/a/art_pyc2jjxzpfrb3ave)

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

Versions

The registry has recorded 5 versions of this file, between Aug 17, 2026 and Sep 23, 2026, about one every 9 days.

versioncommittedcommitsizeaudit
git:20260923.68c095e latest2026-09-23 68c095e 7,128 BB view · diff
git:20260923.cf348bd2026-09-23 cf348bd 7,190 BB view · diff
git:20260821.2a410312026-08-21 2a41031 6,442 BB view · diff
git:20260820.950107c2026-08-20 950107c 4,944 BB view · diff
git:20260817.5b93a212026-08-17 5b93a21 4,964 BB view

Audit of the latest version

B  16 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • fail: No script tag (matched: <script)
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (7128 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

Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.

Source

GitHub

kensaurus/cursor-kenji · 9 stars · license MIT · pushed 2026-10-19 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_pyc2jjxzpfrb3ave
GET https://markdownregistry.com/api/v1/resolve?ref=kensaurus/cursor-kenji/iterate-agent-harness
GET https://markdownregistry.com/api/v1/blob/6dfab1a7b26c012632742ddcb422846ab3b5c47de7883e991795b4ed3409b2fe

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: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from kensaurus/cursor-kenji

approved-plan-execution Cursor rule
kensaurus/cursor-kenji · .cursor/rules/approved-plan-execution.mdc · Guard execution of an approved plan or burndown: forbid reward hacking and feature deletion; require checkpoints…
git:20260923.68c095e · audit A · 9 stars
skill-workflows Cursor rule
kensaurus/cursor-kenji · .cursor/rules/skill-workflows.mdc · Skill routing where a name alone misleads: multi-phase requests, closure modes (complete-everything, burndown-full)…
git:20261001.3fce9fb · audit A · 9 stars
verification-before-completion Cursor rule
kensaurus/cursor-kenji · .cursor/rules/verification-before-completion.mdc · Require fresh, scope-appropriate evidence before any completion claim. Distinguishes implemented, locally verified…
git:20260923.cf348bd · audit A · 9 stars
AGENTS.md AGENTS.md file
kensaurus/cursor-kenji · AGENTS.md
git:20261001.3fce9fb · audit A · 9 stars
llms.txt llms.txt file
kensaurus/cursor-kenji · llms.txt
git:20261009.3771283 · audit A · 9 stars
babysit SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/babysit/SKILL.md · Keep an already-open PR merge-ready: triage comments, resolve clear conflicts, fix CI. Do not merge. Dirty tree →…
git:20260923.68c095e · audit A · 9 stars
canvas SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/canvas/SKILL.md · Create a live React canvas beside chat for standalone analytical artifacts that benefit from visual layout…
git:20260923.cf348bd · audit A · 9 stars
create-hook SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/create-hook/SKILL.md · Create Cursor hooks. Use to create a hook, write hooks.json, add hook scripts, or automate behavior around agent events.
git:20260923.cf348bd · audit A · 9 stars
create-rule SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/create-rule/SKILL.md · Create Cursor rules for persistent AI guidance. Use when user wants to create a rule, add coding standards, project…
git:20260923.cf348bd · audit A · 9 stars
create-skill SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/create-skill/SKILL.md · Guide users through creating effective Agent Skills for Cursor. Use when user wants to create, write, update, or debug…
git:20261001.3fce9fb · audit A · 9 stars
create-subagent SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/create-subagent/SKILL.md · Create custom subagents for specialized AI tasks. Use when the user wants to create a new type of subagent, set up…
git:20260923.68c095e · audit A · 9 stars
migrate-to-skills SKILL.md skill
kensaurus/cursor-kenji · skills-cursor/migrate-to-skills/SKILL.md · Convert 'Applied intelligently' Cursor rules (.cursor/rules/*.mdc) and slash commands (.cursor/commands/*.md) to Agent…
git:20260923.68c095e · audit A · 9 stars

Every file in kensaurus/cursor-kenji

Other files named iterate-agent-harness

iterate-agent-harness SKILL.md skill
kensaurus/skills · skills/iterate-agent-harness/SKILL.md · Turn an agent failure (premature stop, false completion, gamed check, missed file) into a rule, hook, or CI guard with…
git:20260923.68c095e · audit B · 9 stars

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