Home / devantler-tech / agent-skills · agent-improvement/SKILL.md · GitHub

agent-improvement skillA

agent-improvement is agent-read markdown (skill) from devantler-tech/agent-skills: How a meta-engineer improves an autonomous AI engineer from the OUTSIDE — mining the agent's own operational telemetry across many runs and, where a deployment runs more than one instance, across all of them. Scores the agent on reliability, safety, efficiency, outcome throughput, quality, coordination, currency, and prioritization/flow, diagnoses root causes from measured patterns, ships the highest-value fix with its evidence and a reversible audit trail, then verifies the targeted metric actu.

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# Agent-improvement loop

An autonomous AI engineer improves the products it owns. This skill is how something improves **the
engineer itself**, from the outside, on evidence the agent generated by running.

**It complements a `self-improvement` skill rather than replacing it.** Self-improvement is a single
run reflecting on its own memory, on a slow distil cadence. This is an **external observer** across the
whole session corpus — and, in a deployment running more than one instance, across all of them at once.
The observer's own Agent Improver sessions are part of that corpus too: it improves the Engineer's
execution plane and its own observation plane as separately measured subjects. That vantage point is
what makes it see things no single run can:

- failures that recur across hundreds of runs but look like one-offs from inside any single run;
- waste that feels normal from inside a run (a wait, a retry, a re-read) and is obvious in aggregate;
- **divergence between sibling instances** of the same brain;
- **drift between what the agent's bootstrap/loader says and what its contract says** — the loaders are
…

Read the whole file at its exact version.

How to install

Latest version
mdr add devantler-tech/agent-skills/agent-improvement@git:20260924.2edcf23
Exact content
mdr add devantler-tech/agent-skills/agent-improvement@sha256:9aabde439a254846

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_inh5uyhe2wbohvor.svg)](https://markdownregistry.com/a/art_inh5uyhe2wbohvor)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260924.2edcf23 latest2026-09-24 2edcf23 40,007 BA view · diff
git:20260912.9d434db2026-09-12 9d434db 38,192 BA view · diff
git:20260908.cfeec832026-09-08 cfeec83 37,473 BA view · diff
git:20260903.09744842026-09-03 0974484 36,066 BA view · diff
git:20260831.92580382026-08-31 9258038 34,979 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (40007 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

devantler-tech/agent-skills · 3 stars · license Apache-2.0 · pushed 2026-09-24 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_inh5uyhe2wbohvor
GET https://markdownregistry.com/api/v1/resolve?ref=devantler-tech/agent-skills/agent-improvement
GET https://markdownregistry.com/api/v1/blob/9aabde439a25484636e8863d5cc6dd8ee601068c08b37f7e8e90f089669c264e

Your agent 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.

More from devantler-tech/agent-skills

AGENTS.md agents
devantler-tech/agent-skills · AGENTS.md
git:20260924.2edcf23 · audit A · 3 stars
CLAUDE.md claude
devantler-tech/agent-skills · CLAUDE.md
git:20260531.a44ce0b · audit B · 3 stars
agent-instructions skill
devantler-tech/agent-skills · agent-instructions/SKILL.md · Architect a repository's AI-agent instruction files so one canonical source drives every tool without drift — AGENTS.md…
git:20260711.658f1d1 · audit A · 3 stars
allowed-stack-guardrail skill
devantler-tech/agent-skills · allowed-stack-guardrail/SKILL.md · Before agreeing to build anything for a non-technical user, check the need against the consuming deployment's "## Stack…
git:20260712.7b9904e · audit A · 3 stars
conventional-release skill
devantler-tech/agent-skills · conventional-release/SKILL.md · Ship changes through a squash-merge + Conventional Commits release pipeline where the PR title becomes the release note…
git:20260628.69b72ea · audit A · 3 stars
gitops-tenant-onboarding skill
devantler-tech/agent-skills · gitops-tenant-onboarding/SKILL.md · Onboard a new tenant application onto a Flux-based GitOps platform: scaffold the tenant repo's deploy manifests, wire…
git:20260615.7099ba9 · audit A · 3 stars
jargon-free-voice skill
devantler-tech/agent-skills · jargon-free-voice/SKILL.md · The conversational register for assisting a non-technical person who is building a product by conversation alone: plain…
git:20260712.7b9904e · audit A · 3 stars
needs-stack-mapping skill
devantler-tech/agent-skills · needs-stack-mapping/SKILL.md · Translate a non-technical person's plain-language product needs (an outcome, an audience, a workflow) into the…
git:20260712.7b9904e · audit A · 3 stars
portfolio-maintenance skill
devantler-tech/agent-skills · portfolio-maintenance/SKILL.md · The run loop for an autonomous AI engineer acting as a portfolio's primary engineer — pre-flight, resume in-flight work…
git:20260915.a6a7e76 · audit A · 3 stars
product-engineering skill
devantler-tech/agent-skills · product-engineering/SKILL.md · The ADVANCE playbook for an autonomous AI engineer — how to move a product forward once it is healthy: product strategy…
git:20260813.5f0d79a · audit A · 3 stars
self-improvement skill
devantler-tech/agent-skills · self-improvement/SKILL.md · How an autonomous AI engineer improves its OWN definition (its engineering contract, agent definitions, and skills)…
git:20260813.5f0d79a · audit A · 3 stars
ways-of-working skill
devantler-tech/agent-skills · ways-of-working/SKILL.md · Codifies devantler-tech engineering practices: agent-first development workflow, TDD, CI/CD pipelines, GitHub Flow…
git:20260707.a32da31 · audit A · 3 stars

Every file in devantler-tech/agent-skills

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