erstan-agent-optimizer

erstan-agent-optimizer is a skill for Claude Code, Codex from erstanai/erstan-agent-tools-beta. It costs 72 tokens per session (845 once invoked), scanned A, original, Apache-2.0.

Instructions for improving an existing Erstan Agent using its current workflow and evidence from past runs.

In plain words
What is it for?
Improving tool and Skill routing, context use, orchestration, reliability, and protection against unsafe writes.
Why use it?
They help target reliability and efficiency problems without masking defects in Skills, connectors, or the platform.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present.

Part of the erstan-beta plugin — 7 skills, 1 MCP server shipped together

Good fit Improving tool and Skill routing, context use, orchestration, reliability, and protection against unsafe writes.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/erstanai/erstan-agent-tools-beta/erstan-agent-optimizer
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add erstanai/erstan-agent-tools-beta --skill erstan-agent-optimizer
Clone the repo
git clone --depth 1 https://github.com/erstanai/erstan-agent-tools-beta

Made for: Claude Code, Codex.

Or install erstan-beta, the plugin that ships this one along with the rest of its 7 skills, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for erstan-agent-optimizer

README.md
[![agentmods](https://agentmods.dev/badge/skills/erstanai/erstan-agent-tools-beta/erstan-agent-optimizer.svg)](https://agentmods.dev/skills/erstanai/erstan-agent-tools-beta/erstan-agent-optimizer)
Your own site
<a href="https://agentmods.dev/skills/erstanai/erstan-agent-tools-beta/erstan-agent-optimizer"><img src="https://agentmods.dev/badge/skills/erstanai/erstan-agent-tools-beta/erstan-agent-optimizer.svg" alt="Measured on agentmods" height="20"></a>
Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 845 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00072 $0.00845
Opus 5 $0.00036 $0.00423
Sonnet 5 $0.00014 $0.00169
Haiku 4.5 $0.00007 $0.00085

Measured 7d ago against content hash 73b3623c620b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

erstan-agent-optimizer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/erstan-beta/skills/erstan-agent-optimizer/SKILL.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Erstan Agent optimizer

Produce the smallest evidence-backed Agent candidate that improves measured behavior without hiding Skill, connector, or platform defects inside prompt text. Optimization is proposal-only unless the user separately authorizes a saved update, live preview, or publication.

Required protocol

Read the Agent optimization protocol before proposing a candidate or changing an Agent.

Workflow

  1. Establish the requested stage: analysis, local candidate, saved update, live preview, or publication. Treat each later stage as separate authority.
  2. Call get_agent_builder_guide, the relevant detailed get_agent_node_catalog entries, list_agent_tools, and list_agent_skills. Retrieve the complete current Agent with get_agent and record its ID, current/published versions, opaque revision, graph, policies, bindings, and unknown fields.
  3. Correlate diagnostic evidence by Agent ID and executed version. Use get_run plus every get_run_trace page when accessible. When the user supplies an authorized local diagnostics export, inspect its manifest, traces, transcript, and artifact availability without copying sensitive payloads into the report. A different or absent version supports a drift finding, not a claim about the current graph.
  4. Retrieve every bound workspace Skill with get_agent_skill when readable. Record system:<key> or inaccessible Skill packages as evidence gaps. Inspecting a Skill does not authorize changing it.
  5. Classify every finding as Agent, Skill, platform, connector/provider, or a split boundary. Change the Agent only for graph topology, node roles, instructions, policies, bindings, approval design, and orchestration it owns.
  6. Assemble a complete candidate graph and a structured baseline-to-candidate diff. Tie each changed field to direct evidence and an expected observable outcome. Preserve unrelated metadata, positions, nodes, edges, opaque fields, approval controls, stable identities, and safe retry behavior.
  7. Call validate_agent against the candidate when the connected app permits non-mutating validation. Stop at a candidate and validation report unless the user explicitly asked for a saved update.
  8. For an authorized update, re-read the Agent, reconcile version drift, and call update_agent with the latest revision and a new stable idempotency key. Do not test the draft or publish it under the update authorization.
  9. Call test_agent only after explicit live-preview authorization. Real integrations and side effects remain possible. For writes, independently reconcile effects by stable business identity.
  10. Call publish_agent only after explicit publication approval for the exact validated revision. Report the retained published version after a re-read.

Read the full file on GitHub · 74 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 7d ago First seen · 74 lines · 72 tokens per session scan A 73b3623c620b

Subscribe to this mod's changes

erstan-agent-optimizer is a skill published in the GitHub repository erstanai/erstan-agent-tools-beta (0 stars, last pushed 9d ago), licensed Apache-2.0. It adds 72 tokens to every session and 845 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

haiku

When writing a haiku for this bot, follow these conventions.

agno-agi/agno · 0 tokens

deploy-docker-compose

Run the Omnigent server as a Docker compose stack (server + Postgres) on any Docker host — your laptop, a VPS, EC2 by hand, or as the base layer of any container-platform deploy. Invoke when the user wants to build the image, bring up the compose stack, debug the stack on a host they already have, or extend the stack…

omnigent-ai/omnigent · 84 tokens

azure-mgmt-botservice-dotnet

Azure Resource Manager SDK for Bot Service in .NET. Management plane operations for creating and managing Azure Bot resources, channels (Teams, DirectLine, Slack), and connection settings. Triggers: "Bot Service", "BotResource", "Azure Bot", "DirectLine channel", "Teams channel", "bot management .NET", "create bot".

microsoft/skills · 78 tokens

fastapi-router-py

Create FastAPI routers with CRUD operations, authentication dependencies, and proper response models. Use when building REST API endpoints, creating new routes, implementing CRUD operations, or adding authenticated endpoints in FastAPI applications.

microsoft/skills · 46 tokens

dogfood

Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.

callstack/agent-device · 55 tokens