Seamark — AI skill for Claude Code
🛡️ Change control for AI coding agents.
How to install Seamark
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open marcosmatalab/seamark and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Seamark does
🛡️ Change control for AI coding agents. See what Claude Code, Cursor, Codex and Gemini CLI can do in your repo (hooks, MCP servers, permissions) and what every pull request changes. Offline CLI, GitHub Action, SARIF.
Alternatives in AI
- Memory Forge Rs — Stop resetting satisfying AI chats — edit the memory instead 474 ★
- HarnessKit — More than a skill manager — manage skills, MCP servers, plugins, hooks, CLIs, configs, memory & rules across e 416 ★
- New Crud — Generate full CRUD (model, controller, service, views, routes, permissions, hooks) for a module 407 ★
README
⚓ Seamark
**Change control for what your AI agents can do.**
Seamark reads the configuration your coding agents load, resolves what it actually lets them do, and tells you exactly what changed between two commits.
[](https://github.com/marcosmatalab/seamark/actions/workflows/ci.yml) [](pyproject.toml) [](LICENSE) [](#-verify-this-in-60-seconds) [](#in-a-pull-request-the-action-and-the-pre-commit-hook) [](#in-a-pull-request-the-action-and-the-pre-commit-hook)
**Seamark 3.0.0** · Apache-2.0 · one runtime dependency · offline, no telemetry, no account
**[Español](README.es.md)** · [What it does](#-what-it-does-in-plain-words) · [Quickstart](#-quickstart) · [Trade-offs](#-design-trade-offs) · [Commands](docs/COMMANDS.md) · [Design](docs/DESIGN.md)
pip install git+https://github.com/marcosmatalab/seamark && seamark check💡 What it does, in plain words
AI coding agents (Claude Code, Cursor, Codex CLI, Gemini CLI) read configuration from your repository **before you type anything**: hooks that run commands at session start, MCP servers they connect to, permissions, sandbox policy, instruction files. One pull request can quietly give an agent new powers, and no single file tells you what the agent is allowed to do.
**Seamark is the diff for those powers.** It answers three questions:
| Question | Command | |
|---|---|---|
| 🔍 | What can an agent do in this repository right now? | seamark check |
| 🔀 | *What did this pull reque |
Related Skills
AI Harness Character Creator
AgentDeck: one local GUI to configure every AI coding agent. Tune behavior, permissions, token usage and cost,
Propr
Open source, self-hosted GitHub orchestration for AI coding agents. Runs Claude Code, Codex, Antigravity, Open
Truedone
Checks whether a pull request's tests prove its change: red without it, green with it. Flags skipped, focused
Code Agent
An AI Agent that operates on GitHub Actions. By using this action, you can directly invoke Coding Agent CLI fr
Sourceant
Persistent memory for AI coding agents: decisions, rules, API contracts, and system topology served over MCP t
AI Config Sync Manager
Bidirectional sync for Claude Code and Codex — instructions, skills, agents, MCP servers, hooks, permissions.
Related Agents
Agentic Auditor
PYA specialist spawned by the /pya orchestrator. Audits AI agent and LLM surfaces in an untrusted repository:
Claude Code Infra
Use PROACTIVELY and automatically — do not wait to be asked — to configure Claude Code itself: sub-agents, ski
Ccxt PR Reviewer
End-to-end review of a CCXT pull request. Reads the diff, transpiles and builds in all six languages (TS/JS, P
