portal-router
Resolve and load any agent capability across the whole fleet from one place. Given an intent, finds the matching skill, command, or agent across Claude, Codex, Gemini, .agents, a-i--skills, OpenCode, and OpenClaw — then loads it through a single portal. Use for cross-agent skill discovery, "which agent has a skill for X", capability routing, or "find a skill that does Y" when capabilities are spread across multiple agent ecosystems.
What this skill does
# portal-router — cross-agent capability resolution
When capabilities are spread across many agent ecosystems (Claude, Codex, Gemini,
OpenCode, OpenClaw, a shared `.agents` pool, a skills distribution), the expensive
question is no longer "how do I do X" but **"which of my agents already has a skill
for X, and how do I load it?"** `portal-router` answers that from one place.
It treats every ecosystem as a capability source, normalizes their heterogeneous
unit shapes into one record, ranks matches against an intent, and loads the chosen
unit. It is the skill an orchestrator reaches for *before* authoring a new skill —
to check whether the fleet already owns the capability.
## When to use
- "Which agent has a skill for PDF extraction / k8s / data cleaning?"
- "Find me a capability that does Y" — across all agents, not just the current one.
- Before forging a new skill: confirm the fleet doesn't already have one.
- Routing a task to whichever agent owns the right tool.
## The load-path
The reference implementation is the operator's `~/_arms/arm` portal (the `_arms`
cross-agent skill ecosystem). It exposes:
```sh
arm find <query…> # rank capabilities across ALL ecosystems by relevance
arm list [ecosystem] # the portal index (computed live; never stale)
arm show <eco> <name> # LOAD one capability through the portal (prints its body)
arm status # portal health
```
Example — find a capability regardless of which agent owns it, then load it:
```sh
arm find pdf extraction
arm show claude pdf-processing # capability loaded through the portal into context
```
## How it resolves (the engine)
`references/resolve.py` (stdlib-only Python — no dependencies) is the portable
engine. It:
1. Discovers every ecosystem's skill home (following symlinks — `find -L` /
`followlinks=True`; the mirror entries are symlinks, and without following them
the portal reads as empty — the single load-bearing detail).
2. Normalizes heterogeneous units — Claude/Codex/Gemini `SKILL.md` dirs with YAML
frontmatter, OpenCode flat command `.md` (name = filename), a-i--skills
`*.skill`/category dirs — into one `{ecosystem, type, name, description, path,
invoke}` record.
3. Ranks matches (name > triggers > description > path) and, on `show`, prints the
unit plus the exact invoke hint for that agent's own loader.
Point it at any directory of ecosystems (default `~/_arms/mirror/` + `~/_arms/skills/`)
and it indexes them live. The index is computed each call from the sources, so it
can never drift from reality — there is no stored index to stale.
## Honest about reach
For same-agent units (e.g. Claude loading a Claude `SKILL.md`), `show` is a real
context-load. For other agents it prints the unit and the precise invoke hint for
*that* agent's loader — the portal routes you to the door; each agent still opens
its own. Cross-agent *execution* is the orchestrator's job (see `agent-swarm-orchestrator`);
`portal-router` is the discovery-and-routing layer beneath it.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.