skillopt

An optimization loop called SkillOpt that improves an agent capability over repeated epochs and mini-batches. An epoch is one pass through a group of training tasks, and its edit budget controls how much text or code may change in each step.

In plain words
What is it for?
Use it to optimize a prompt, tool surface, or skill package through a single sequence of edits, validation scores, and accept-or-reject decisions.
Why use it?
It organizes repeated improvement attempts with a shrinking edit budget and an extra consolidation step at the end of each epoch.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/skillberry-ai/cap-evolve/skillopt
Any agent
npx skills add skillberry-ai/cap-evolve --skill skillopt
Clone the repo
git clone --depth 1 https://github.com/skillberry-ai/cap-evolve

Made for: Claude Code, Codex.

Per session 118 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,480 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00118 $0.02480
Opus 5 $0.00059 $0.01240
Sonnet 5 $0.00024 $0.00496
Haiku 4.5 $0.00012 $0.00248

Measured 3d ago against content hash cb8a56890c34, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skillopt 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 3d ago.

The scan reads SKILL.md. This mod also ships 4 executable files (scripts/_bootstrap.py, scripts/abstract.py, scripts/check.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/algorithms/skillopt/SKILL.md · 152 lines

How it starts

The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.

skillopt — annealed single-lineage climb (epochs × mini-batches)

SkillOpt (arXiv:2605.23904, Executive Strategy for Self-Evolving Agent Skills) organizes a hill-climb into epochs × mini-batches under a decaying integer edit budget. The name is the paper's; the algorithm edits whatever the selected capability owns — a prompt, a tool surface, a skill package — and never assumes which.

Read the shared step first, then this file. Parent materialization, the optimizer call, the val evaluation, the significance gate, accept/reject, snapshot/best, the memory and handover files: all of that is harness.run_step, documented once in algorithms/hill-climb/SKILL.md § "One iteration, end to end" and algorithms/hill-climb/references/run-step.md. This file states only what SkillOpt does differently.

Know the bound before reaching for this algorithm: run_step lets a caller vary exactly two things — parent_dir and instructions. SkillOpt pins parent_dir to the current best, identical to hill-climb, so everything novel lives in the instructions string plus the choice to run one extra step per epoch. It is prompt shaping and step scheduling, not a different search.

What SkillOpt does differently

  1. A decaying integer edit budget L. lr_schedule.build_schedule emits one integer per step over constant | linear | cosine, clamped to [--min-edit-budget, --edit-budget] (core/cap_evolve/lr_schedule.py:42-55). L is stated to the optimizer in prose — "at most L bounded edits" — and is never mechanically enforced. See the next section before you tune it.
  2. A per-epoch rejected-edit list. Each reject appends its candidate id and val Δ, and the next step's prompt asks the optimizer to avoid them (skillopt.py:120-125, :330-335). It carries no description of what the rejected edit changed, so treat it as a weak signal — the run-global LEDGER.md that run_step already injects names the tasks each prior edit broke and fixed, which is strictly more useful.
  3. One extra gated step per epoch boundary (from epoch 2). It compares the epoch-start candidate against the current best, buckets tasks as regressed / persistent-failure / stable-success, and asks for a consolidating edit that fixes regressions without breaking the stable passes. It goes through the same run_step and the same val gate — it is never force-accepted (skillopt.py:493-499). Disable with --no-slow-update. A fourth bucket, improved, is computed and logged but is not exclusive with the others and never reaches the prompt (skillopt.py:184-191, :139-166).

Read the full file on GitHub · 152 lines

Files

What ships with it

6 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. 3d ago First seen · 152 lines · 118 tokens per session scan A cb8a56890c34

Subscribe to this mod's changes

skillopt is a skill published in the GitHub repository skillberry-ai/cap-evolve (47 stars, last pushed 3d ago), licensed Apache-2.0. It adds 118 tokens to every session and 2,480 once invoked, about $0.0006 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 2026-08-30.