hephaestus-storm

hephaestus-storm is a skill for Claude Code, Codex from agentlas-ai/Agentlas-OS. It costs 132 tokens per session (2,408 once invoked), scanned C, original, Apache-2.0.

A force-robust execution loop for driving a goal through real Agentlas specialists until the work is verified. It breaks work into dependency-ordered packets, runs each packet through a controlled goal loop, and requires evidence before reporting success.

In plain words
What is it for?
Use it to route a goal to specialists, build and run a pipeline of dependent work packets, include available coding-agent sessions, and verify the final result before declaring success.
Why use it?
It addresses the common problem of one-shot automation stopping early, losing dependencies, or claiming completion without proof. It keeps the execution process and verification rules under the Agentlas Core harness.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Claude Code; installed under .agents/ (shared by several agents); mentions Codex.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is ./bin/hephaestus.

Good fit Use it to route a goal to specialists, build and run a pipeline of dependent work packets, include available coding-agent sessions, and verify the final result before declaring success.

Compare 6 skills from other repositories ↓
About the project

Agentlas OS is a local-first system for creating, storing, borrowing, and running specialist AI agents and temporary agent teams through supported hosts and models. It serves people who want reusable agents that remain available across computers and model workspaces, and the catalogue contains its skills, commands, hooks, agents, instructions, plugin, and rule.

agentlas-ai/Agentlas-OS · 1,105 stars · on GitHub · agentlas.cloud

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/agentlas-ai/Agentlas-OS
agentmods
npx agentmods add skills/agentlas-ai/agentlas-os/hephaestus-storm

Made for: Claude Code, Codex.

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 hephaestus-storm

README.md
[![agentmods](https://agentmods.dev/badge/skills/agentlas-ai/agentlas-os/hephaestus-storm/github.svg)](https://agentmods.dev/skills/agentlas-ai/agentlas-os/hephaestus-storm)
Your own site
<a href="https://agentmods.dev/skills/agentlas-ai/agentlas-os/hephaestus-storm"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-os/hephaestus-storm/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for hephaestus-storm

Your own site · 80×15
<a href="https://agentmods.dev/skills/agentlas-ai/agentlas-os/hephaestus-storm"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-os/hephaestus-storm.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 132 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,408 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
SkillSpector: 1 finding, up to low

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • low Supply Chain · line 57
    Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
    Fix: Avoid downloading and executing remote scripts. Use trusted packages from PyPI/npm. If remote fetch is required, verify checksums and use HTTPS.
How audits are shown
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 $0.00132 $0.02408
Opus 5 $0.00066 $0.01204
Sonnet 5 $0.00026 $0.00482
Haiku 4.5 $0.00013 $0.00241

Measured 9d ago against content hash c1d4444a8c48, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade C, and why

hephaestus-storm scanned grade C with 2 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 9d 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

`curl -fsSL https://raw.githubusercontent.com/agentlas-ai/Agentlas-OS/main/scripts/install-all-runtimes.sh | bash`

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

`curl -fsSL https://raw.githubusercontent.com/agentlas-ai/Agentlas-OS/main/scripts/install-all-runtimes.sh | bash`
.agents/skills/hephaestus-storm/SKILL.md · 181 lines

How it starts

The opening of the file, as written. The whole thing — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Update fallback: 자동 업데이트가 안 되면 hephaestus update를 한 번 실행하세요. 업데이트하지 않아도 현재 버전 명령은 그대로 동작합니다.

Hephaestus Stormbreaker Loop

Drive a goal through the Stormbreaker Loop — Hephaestus' force-robust, verifier-first execution loop. Unlike a one-shot answer or a generic parallel fan-out, Stormbreaker routes the goal to real Agentlas specialists, structures the work as a dependency-ordered pipeline fabric, drives each work packet as a hardened goal loop (it does not stall, run away, or claim false success), and refuses to report success without evidence. Never guess an agent yourself when this skill is active — the router or Hub decides the workforce.

Core-owned Goal + UltraCode harness

Every hep-storm result includes execution_harness. Before planning or executing any packet, apply execution_harness.system_prompt verbatim and retain its prompt_sha256 in the goal ledger. This adapter must never redefine, summarize, or replace Goal mode or UltraCode mode with host-local wording. The adapter owns invocation only; Agentlas Core owns the execution protocol.

If the host exposes live Codex, Claude Code, Gemini, local-model, or other sessions, provide their JSON array through AGENTLAS_SESSION_INVENTORY or --session-inventory. If it does not, accept Core's explicit host:primary fallback; never invent a model ID or claim unavailable parallel workers.

With no external executor, the runner intentionally returns status: materialized and final_gate.can_report_success: false. That is the host's signal to execute the returned packets with its native tools; it is not a failure and must never be rewritten as completion.

1. Resolve the runner

Run this resolution in a shell and use the first hit:

RUNNER=""
for c in \
  "$HOME/.agentlas/runtime/current/bin/hephaestus" \
  ./bin/hephaestus
do [ -x "$c" ] && RUNNER="$c" && break; done
if [ -z "$RUNNER" ]; then
  for cache in \
    "$HOME/.claude/plugins/cache/agentlas-core-engine/hephaestus" \
    "$HOME/.codex/plugins/cache/agentlas-core-engine/hephaestus"; do
    newest="$(ls -d "$cache"/*/bin/hephaestus 2>/dev/null | sort -V | tail -1)"
    [ -n "$newest" ] && [ -x "$newest" ] && RUNNER="$newest" && break
  done
fi

Read the full file on GitHub · 181 lines

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. 9d ago First seen · 181 lines · 132 tokens per session scan C c1d4444a8c48

Subscribe to this mod's changes

hephaestus-storm is a skill published in the GitHub repository agentlas-ai/Agentlas-OS (1,105 stars, last pushed 3d ago), licensed Apache-2.0. It adds 132 tokens to every session and 2,408 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

agentfield-use

Whenever you have a discrete task to perform — one the user delegated, or one that arose inside your own work — check FIRST whether an installed AgentField agent covers it, and offload to it by default when one does. Coverage, not task size, is the test: even a small job goes to a covering agent. The check is cheap …

Agent-Field/agentfield · 259 tokens

agent-optimization

Improve an Agent State through versioned scores and score-linked Traces from a frozen Benchmark.

Prism-Shadow/penguin-harness · 22 tokens

agent-evaluation

Run one specified Test Agent on one specified Benchmark Case exactly once, privately score that execution, and return one protocol result.

Prism-Shadow/penguin-harness · 28 tokens

quality-loop

Use this workflow recipe when a draft, plan, proposal, or other deliverable should be independently reviewed and revised until it satisfies explicit quality criteria.

trpc-group/trpc-agent-go · 32 tokens

cli-skill-design

Design a co CLI surface and its SKILL.md together so an agent can drive it without guessing — every command ends by naming the next one, --help lists everything, and every failure says what to run instead. Use when adding a new CLI command group, writing or rewriting a SKILL.md for one, or auditing an existing one.

openonion/connectonion · 79 tokens

agent-identity

Verify an agent's live address and billing account before sharing a chat link, diagnosing a deploy mismatch, or deciding why two hosts share an identity.

openonion/connectonion · 33 tokens