steering

steering is a command for Claude Code from growilabs/growi-mcp-server. It costs 12 tokens per session (408 once invoked), scanned A, a copy of steering, MIT.

A project-knowledge manager for `.kiro/steering/`, a folder of files that records product goals, technology choices, and project structure.

In plain words
What is it for?
Use it to create the initial knowledge files or update them after the project changes.
Why use it?
It keeps important project decisions available to coding agents and highlights when the code no longer matches them.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions subagents.

Part of the mcp-client-skills plugin — 2 skills, 12 commands shipped together

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 commands/growilabs/growi-mcp-server/steering
Clone the repo
git clone --depth 1 https://github.com/growilabs/growi-mcp-server

Made for: Claude Code.

Or install mcp-client-skills, the plugin that ships this one along with the rest of its 2 skills, 12 commands.

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 steering

README.md
[![agentmods](https://agentmods.dev/badge/commands/growilabs/growi-mcp-server/steering.svg)](https://agentmods.dev/commands/growilabs/growi-mcp-server/steering)
Your own site
<a href="https://agentmods.dev/commands/growilabs/growi-mcp-server/steering"><img src="https://agentmods.dev/badge/commands/growilabs/growi-mcp-server/steering.svg" alt="Measured on agentmods" height="20"></a>
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 408 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 92% copy Near-identical to another mod 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 $0.00012 $0.00408
Opus 5 $0.00006 $0.00204
Sonnet 5 $0.00002 $0.00082
Haiku 4.5 $0.00001 $0.00041

Measured 6d ago against content hash 7e1a8949ad60, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

steering 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 6d 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.

Origin

This is a copy

92% identical to steering — 16 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/commands/kiro/steering.md · 63 lines

What it actually says

Kiro Steering Management

Mode Detection

Perform detection before invoking Subagent:

Check .kiro/steering/ status:

  • Bootstrap Mode: Empty OR missing core files (product.md, tech.md, structure.md)
  • Sync Mode: All core files exist

Use Glob to check for existing steering files.

Invoke Subagent

Delegate steering management to steering-agent:

Use the Task tool to invoke the Subagent with file path patterns:

Task(
  subagent_type="steering-agent",
  description="Manage steering files",
  prompt="""
Mode: {bootstrap or sync based on detection}

File patterns to read:
- .kiro/steering/*.md (if sync mode)
- .kiro/settings/templates/steering/*.md
- .kiro/settings/rules/steering-principles.md

JIT Strategy: Fetch codebase files when needed, not upfront
"""
)

Display Result

Show Subagent summary to user:

Bootstrap:

  • Generated steering files: product.md, tech.md, structure.md
  • Review and approve as Source of Truth

Sync:

  • Updated steering files
  • Code drift warnings
  • Recommendations for custom steering

Notes

  • All .kiro/steering/*.md loaded as project memory
  • Templates and principles are external for customization
  • Focus on patterns, not catalogs
  • "Golden Rule": New code following patterns shouldn't require steering updates
  • Avoid documenting agent-specific tooling directories (e.g. .cursor/, .gemini/, .claude/)
  • .kiro/settings/ content should NOT be documented in steering files (settings are metadata, not project knowledge)
  • Light references to .kiro/specs/ and .kiro/steering/ are acceptable; avoid other .kiro/ directories
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. 6d ago First seen · 63 lines · 12 tokens per session scan A 7e1a8949ad60

Subscribe to this mod's changes

steering is a command published in the GitHub repository growilabs/growi-mcp-server (19 stars, last pushed 1mo ago), licensed MIT. It adds 12 tokens to every session and 408 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to steering, differing in 16 lines, and is treated as a copy.