AgentHydra: Command for Claude Code

.claude/commands/hydra-check.md

hydra-check is a command for Claude Code from LunarWerxs/AgentHydra. It costs 27 tokens per session (355 once invoked), scanned A, original, MIT.

A command that runs the repository’s complete pre-push checks locally: formatting, type checks, tests, guardrails, and the same workflow steps used by its continuous-integration checks.

In plain words
What is it for?
Use it before pushing changes to check the Bun project, run server and web type checks, execute tests, and run the repository’s workflow-based local CI checks.
Why use it?
It shows the real pass or fail result before code is pushed and helps distinguish an actual bug from another agent’s unfinished edits.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution. Also seen: reads .claude/ paths.

This is LunarWerxs/AgentHydra's own configuration. It tells Claude Code how to work on AgentHydra itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything AgentHydra configures →

Reuse

Borrowing it

Nothing to install: this file belongs to LunarWerxs/AgentHydra. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/LunarWerxs/AgentHydra/main/.claude/commands/hydra-check.md
Clone the repo
git clone --depth 1 https://github.com/LunarWerxs/AgentHydra

Made for: Claude Code.

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 hydra-check

README.md
[![agentmods](https://agentmods.dev/badge/commands/lunarwerxs/agenthydra/hydra-check/github.svg)](https://agentmods.dev/commands/lunarwerxs/agenthydra/hydra-check)
Your own site
<a href="https://agentmods.dev/commands/lunarwerxs/agenthydra/hydra-check"><img src="https://agentmods.dev/badge/commands/lunarwerxs/agenthydra/hydra-check/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 hydra-check

Your own site · 80×15
<a href="https://agentmods.dev/commands/lunarwerxs/agenthydra/hydra-check"><img src="https://agentmods.dev/badge/commands/lunarwerxs/agenthydra/hydra-check.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 355 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 $0.00027 $0.00355
Opus 5 $0.00014 $0.00178
Sonnet 5 $0.00005 $0.00071
Haiku 4.5 $0.00003 $0.00036

Measured 9d ago against content hash 50bc07d6ca85, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

hydra-check 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 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.

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.

.claude/commands/hydra-check.md · 25 lines

What it actually says

Run every gate this repo has, in order, and report the real verdict. Do not push, and do not declare it green on a partial run. Extra instructions, if any: $ARGUMENTS

  1. bunx biome check --write . - format and lint.
  2. bun run --cwd server typecheck and bun run --cwd web typecheck.
  3. bun run test - the full suite (parallel, four workers; see the workflow comment for why CI itself stays serial).
  4. python ~/.claude/tools/localci.py - the one that matters: it reads this repo's own .github/workflows/*.yml and runs those steps here, so the workflow file IS the gate list and there is no second copy to drift. Includes every guardrail check.

Rules:

  • A red is not automatically yours. These trees run many concurrent agents; before believing a failure, check git status --short and the mtimes of the named files - someone else's half-finished edit fails exactly like a bug. Say so rather than "fixing" work in progress.
  • A green here is ONE leg. localci covers Windows (and Linux in a container with --docker); macOS is always GitHub's, and it says so.
  • If anything fails, fix the cause. Never weaken a test or a guardrail to get green - a check that cries wolf gets ignored, and then it misses the real thing.
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 · 25 lines · 27 tokens per session scan A 50bc07d6ca85

Subscribe to this mod's changes

hydra-check is a command published in the GitHub repository LunarWerxs/AgentHydra (43 stars, last pushed yesterday), licensed MIT. It adds 27 tokens to every session and 355 once invoked, about $0.0001 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.