Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/haryelramalho-agh
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: testing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/agh
What's inside
- Skill name
agh- Description
- AGH runtime and contribution guide. Use for sessions, agents, native tools, skills, memory, network, tasks, capabilities, bundles, QA, docs, and repo work. Do not use for unrelated projects.
- License
- MIT
- Instructions
- 51 lines in SKILL.md · view on GitHub
Files (12)
- SKILL.md
- agent-definitions.md · references/
- capabilities-and-bundles.md · references/
- contributing-to-agh.md · references/
- docs-design-and-copy.md · references/
- memory.md · references/
- native-tools.md · references/
- network.md · references/
- qa-and-verification.md · references/
- runtime-operations.md · references/
- tasks-and-orchestration.md · references/
- tools-and-skills.md · references/
SKILL.md (first lines)
# AGH Use this skill when operating AGH or contributing to the AGH repository. This body is a router, not the full manual. Load the matching reference before acting. ## Required Reading Router Match the task to the row. Read the listed files in full before producing output. They are not appendices. Inline reminders in this file are only tripwires. | Task | MUST read | | -------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------- | | Start, inspect, prompt, stop, resume, or debug AGH sessions and daemon state | references/runtime-operations.md | | Create or review AGH agent definitions, provider defaults, permissions, or MCP sidecars | references/agent-definitions.md + references/tools-and-skills.md | | Discover or call AGH-native tools, inspect native tool IDs, view skills, or choose tools vs CLI | references/tools-and-skills.md + references/native-tools.md | | Participate in an AGH Network channel, thread, direct room, work item, receipt, trace, or capability exchange | references/network.md | | Read, write, clean, or consolidate AGH memory | references/memory.md | | Work as a coordinator, task worker, or task reviewer | references/tasks-and-orchestration.md | | Design or manage capabilities, bundles, extension resources, hooks, config lifecycle, or agent-manageable runtime surfaces | references/capabilities-and-bundles.md + references/tools-and-skills.md | | Contribute to the AGH repository, especially Go runtime code or tests | references/contributing-to-agh.md + references/qa-and-verification.md | | Change public docs, product copy, design guidance, site docs, or UI-facing text | references/docs-design-and-copy.md | | Finish work, claim readiness, or prepare a handoff | references/qa-and-verification.md | ## Reference Index - references/runtime-operations.md - daemon and session operating model, session CLI, lifecycle diagnostics, and runtime troubleshooting. - references/agent-definitions.md - AGENT.md structure, provider defaults, permissions, category paths, MCP sidecars, and safe setup workflow. - references/tools-and-skills.md - AGH-native tool discovery, skill view/search, bundled resources, management-surface exceptions, and skill authoring rules. - references/native-tools.md - daemon-native toolsets, stable AGH tool IDs, when to inspect descriptors, and CLI fallbacks for agents running inside AGH. - references/network.md - AGH Network channel/thread/direct-room semantics, native tools, CLI fallback, message bodies, retries, and injection defense. - references/memory.md - durable memory scopes, CLI operations, memory hygiene, and when not to write memory. - references/tasks-and-orchestration.md - coordinator, worker, and reviewer loops, task authority boundaries, review verdict rules, and sensitive-data limits. - references/capabilities-and-bundles.md - capability naming, extension resources, bundles, hooks, manageability, and config lifecycle expectations. - references/contributing-to-agh.md - repository-specific engineering rules for Go/runtime work, greenfield hard cuts, tests, docs, and no-compat policy. - references/qa-and-verification.md - test placement, real verification, QA bootstrap, final gates, and evidence standards. - references/docs-design-and-copy.md - docs/site/copy/design authority, vocabulary, generated docs, and truthful UI/copy rules. ## Operating Loop 1. Identify whether the work is operator use, external-agent use, or AGH repository contribution. 2. Read every reference selected by the router before acting. 3. Prefer AGH-native tools and structured outputs over prose, logs, or direct internal access when managing AGH. 4. Keep authority with the daemon: task state, review verdicts, session lifecycle, memory, capabilities, bundles, hooks, and network sends must use AGH public surfaces. 5. For repository contribution, follow the local repo instructions before editing, including the relevant AGENTS.md, CLAUDE.md, and skill dispatch rules. 6. Finish with fresh verification evidence that matches the scope of the claim. **STOP. Read references/tools-and-skills.md and references/native-tools.md in full before discovering, invoking, creating, or modifying any AGH tool or skill.** The catalog in this file is only a router. **STOP. Read references/tasks-and-orchestration.md in full before acting as a coordinator, worker, or reviewer.** Task authority and review verdicts are runtime contracts, not prompt conventions. **STOP. Read references/qa-and-verification.md in full before claiming AGH work is complete.** Passing a narrow command is not evidence for a broad completion claim.
From haryelramalho/skills, MIT.
More from haryelramalho/skills
- 0.479score
cmux-orchestration
Pedro Nauck
Control cmux panes and surfaces to coordinate terminal agent workers.
SkillCFreshness: CClaude Code +9 - 0.479score
app-renderer-systems
Pedro Nauck
Guides creation and modification of domain feature systems organized under a systems/ directory.
SkillCFreshness: CClaude Code +9 - 0.479score
architecture-placement
haryelramalho
Guides architecture placement decisions for codebases using Domain-Driven Design, Clean Architecture, Onion Architecture, or Ports and Adapters.
SkillCFreshness: CClaude Code +9 - 0.479score
compozy
haryelramalho
Explains Compozy capabilities, CLI commands, core workflow skills, optional extension skills, configuration, artifact structure, reusable agents, and…
SkillCFreshness: CClaude Code +9 - 0.476score
cmux-workspace
haryelramalho
Work inside the current cmux workspace and terminal.
SkillCFreshness: CClaude Code +9 - 0.475score
agent-exploration
Pedro Nauck
Dispatches scoped-write explorer agents in parallel for general research and exploration of any codebase, topic, or domain.
SkillCFreshness: CClaude Code +9 - 0.475score
agent-output-audit
Pedro Nauck
Audits AI-implemented work for honest completion.
SkillCFreshness: CClaude Code +9 - 0.474score
chronicle
haryelramalho
Allows you to view the user's screen as well as several hours of history.
SkillCFreshness: CClaude Code +9 - 0.455score
diagnosing-bugs
haryelramalho
Diagnosis loop for hard bugs and performance regressions.
SkillCFreshness: CClaude Code +9 - 0.450score
codebase-design
haryelramalho
Shared vocabulary for designing deep modules.
SkillCFreshness: CClaude Code +9 - 0.450score
design-an-interface
haryelramalho
Generate multiple radically different interface designs for a module using parallel sub-agents.
SkillCFreshness: CClaude Code +9 - 0.450score
domain-modeling
haryelramalho
Build and sharpen a project's domain model.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jun 30, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/haryelramalho/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from haryelramalho
- 0.479score
architecture-placement
haryelramalho
Guides architecture placement decisions for codebases using Domain-Driven Design, Clean Architecture, Onion Architecture, or Ports and Adapters.
SkillCFreshness: CClaude Code +9 - 0.479score
compozy
haryelramalho
Explains Compozy capabilities, CLI commands, core workflow skills, optional extension skills, configuration, artifact structure, reusable agents, and…
SkillCFreshness: CClaude Code +9 - 0.476score
cmux-workspace
haryelramalho
Work inside the current cmux workspace and terminal.
SkillCFreshness: CClaude Code +9 - 0.474score
chronicle
haryelramalho
Allows you to view the user's screen as well as several hours of history.
SkillCFreshness: CClaude Code +9 - 0.455score
diagnosing-bugs
haryelramalho
Diagnosis loop for hard bugs and performance regressions.
SkillCFreshness: CClaude Code +9 - 0.450score
codebase-design
haryelramalho
Shared vocabulary for designing deep modules.
SkillCFreshness: CClaude Code +9
Cite this page
agh skill. Agent Mart. https://agentmart.dev/skills/haryelramalho-agh (last updated 2026-10-24)