New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Routing Card Authoring

by agentlas-ai in agentlas-ai/Agentlas-OS

A skill your agent uses whenever a build emits or repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder, the team builder, and the packager.

Apache-2.0Auto-check passedAI & LLM Engineering

Install Routing Card Authoring

skills CLI
$ npx skills add agentlas-ai/Agentlas-OS --skill routing-card-authoring -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install agentlas-ai/Agentlas-OS routing-card-authoring --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/agentlas-ai/Agentlas-OS.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/routing-card-authoring .claude/skills/routing-card-authoring && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
routing-card-authoring
GitHub stars
1.6k
Token cost
~2.7k tokens
SKILL.md length
1,243 words
Files
1
Skills in repo
53
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses whenever a build emits or repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder, the team builder, and the packager.

  • Works in 2 steps: Only two columns can be compared between… → The real matching is…
  • Repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder
  • SKILL.md covers Why the fields behave the way…, Field table, Checklist a builder can run and Verified end to end
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Routing Card Authoring is an agent skill from agentlas-ai/Agentlas-OS. Use whenever a build emits or repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder, the team builder, and the packager. States what belongs in every field, which fields the hub can actually match on, and which fields silently break matching when a sentence leaks into them.

Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering, covering Building AI agents. The repository describes itself as: Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model. The licence is Apache-2.0.

When your agent uses it

  • Repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder
  • The team builder

Example prompts

  • “/routing-card-authoring”

Workflow steps

2 steps, taken from the first numbered list in SKILL.md.

  1. Only two columns can be compared between a work order and a card.
  2. The real matching is sentence-to-sentence. The slot's task text is

What it can do on your machine

Read from SKILL.md and the folder at commit 46420de. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Routing Card Authoring loads about 2.7k tokens when it runs. Until then it costs about 84 tokens; SKILL.md has 1,243 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~84
When it runs · the whole SKILL.md, loaded when a task matches
~2.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from agentlas-ai/Agentlas-OS at commit 46420de, republished under its Apache-2.0 licence (© agentlas-ai). 1,243 words, ~2,745 tokens.

Download SKILL.mdSave it as .claude/skills/routing-card-authoring/SKILL.md (or your agent's skills folder).
name
routing-card-authoring
description
Use whenever a build emits or repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder, the team builder, and the packager. States what belongs in every field, which fields the hub can actually match on, and which fields silently break matching when a sentence leaks into them.

Routing Card Field Spec — what goes in every field, and where it lands

One card, four builders. 10-single-agent-builder, 20-multi-agent-team-builder, 30-agentlas-packager, and 40-session-agent-builder differ in what they assemble, but the routing card is the same artifact with the same rules in all four. This file is the reference they share, and templates/routing-card.example.json is a complete card that passes schemas/routing-card.schema.json.

Why the fields behave the way they do

Every rule below was measured against the live corpus on 2026-07-30 (492 workforce profiles, 253 stored manifests). The two facts that decide everything else:

  1. Only two columns can be compared between a work order and a card. communities (422 of 492 populated, 105 distinct) and skills (492 of 492, 1,969 distinct). Everything else is either declared by almost nobody (roles 6, tools 10, knowledge 0, forbiddenAuthorities 0), or declared by everyone with the same value (runtimes 14 distinct over 492 profiles, languages exactly 2, modalities exactly 1) — neither can separate candidates.
  2. The real matching is sentence-to-sentence. The slot's task text is compared semantically against the card's summary. A card whose summary names its actual deliverable outranks a card with a perfect id list.

So the card has two jobs, and mixing them is the classic defect: short ids for retrieval, whole sentences for judgement. A sentence that leaks into an id field becomes a slug nothing else can ever match — measured: 8,973 distinct outputs values across the corpus, 0% shared by two agents.

Field table

Columns: what the field is for · what the hub does with it · what to write.

Identity and display
FieldHub useWhat to write
schemaVersionnoneExactly "routing-card/2.0".
card_versionnoneSemver of this card's content, bumped when you edit it.
ididentitylocal/<package-slug> before upload.
canonical_ididentityStable cross-registry id when one exists, else omit.
typehard filter (entityKind)"agent", "team", or "plugin". A team means an orchestrator owns workers; do not label a single worker a team.
name / name_kodisplay + lexical retrievalThe job title a human would search for. Not a product pun.
aliaseslexical retrievalOther phrasings a requester might type, including Korean. Cheap and safe to add.
supersedeslineageIds this card replaces.
The sentences that actually win matches
FieldHub useWhat to write
summary (≤240 chars)semantic ranking against the slot task — the single highest-value fieldOne sentence: what it does, what the requester ends up holding, and the one boundary that matters. Name the deliverable in the words a requester would use. Do not list technologies for their own sake.
summary_kodisplay + Korean retrievalFaithful Korean of summary.
descriptionsemantic ranking2–4 sentences: when to use it, what the deliverable contains, and what it explicitly does not do. This is where "does not implement" or "does not run migrations" belongs.
trigger_examplessemantic ranking (strong)6+ real sentences a requester would actually type, 3 Korean and 3 English. Write the request, not a feature name.
anti_triggersnegative ranking4+ sentences that look adjacent but must NOT route here. This is how a design agent stops absorbing implementation work.
known_failure_caseshonesty, read by the host LLMWhat the agent degrades to when an input is missing. Sentences, whole.
Short-id fields (open semantic concepts — never sentences)
FieldHub useWhat to write
capabilities→ skills column, lexical + semantic retrieval4–8 ids in verb_object snake_case (design_backend_services). Schema enforces the pattern. These become open skill:* graph concepts. Seed aliases may normalize familiar terms, but no finite list owns what an agent is allowed to know or do.
domainslexical retrieval2–5 broad area words.
required_inputs / optional_inputscompatibility displayCanonical interfaces live in contracts/intake.schema.json. Only stable name/id values may project to ontology IDs; descriptions and JSON types never do.
input_notesnot read by matchingFree-text lines explaining each input, for humans and for the executing model.
consumes / producescompatibility displayCanonical interfaces live in contracts/intake.schema.json and contracts/output.schema.json. When present, use short stable kind IDs only.
supported_runtimespackaging metadataDo not emit it as agent identity. Runtime adapters and prepare-time execution policy own compatibility.
required_plugins→ tools column{id, min_permissions} for a facility the worker itself must invoke. Almost no card declares tools; state the requirement vendor-free and keep the package usable without it.
Show full SKILL.md (548 more words)Show less
The workforce block — open-world semantic résumé fields

workforce is defined in schemas/routing-card.schema.json and is what the hub reads. The snapshot agentlas_cloud/workforce/ontology_v1.json (awo:2026-07-15.2) supplies seed aliases, parent relations and compatibility mappings. It is not an allowlist. When no seed concept fits, author a faithful English namespaced ID; the Hub embeds it and connects it to the ontology graph.

FieldHub useWhat to write
workforce.communitiessemantic scope + strong ranking signal1–5 open community:* professional-domain ids. Reuse a seed alias when it fits; otherwise author the real community.
workforce.skillssemantic ranking3–12 open skill:* verb-object capability ids that actually distinguish this agent. They are not a fixed dictionary.
workforce.knowledgerankingknowledge:* ids. You may omit it: the compiler now derives one id per knowledge/*.md file the package ships (file stem → knowledge:<stem>), which is why the column existed with 0 producers until 2026-07-30.
workforce.rolessemantic ranking0–4 open role:* professional-responsibility ids. [] is honest when the card already expresses the work better through its summary and skills.
workforce.languagesdelivery metadata, never an identity/callability gateLanguages the work product can genuinely be delivered in. Listing translation does not count. [] is allowed when unknown.
workforce.modalitiesruntime/input metadata, never an identity/callability gateDeclare only non-text media the agent genuinely consumes or emits. Ordinary text reasoning may use [].
Safety, cost and operations (not matching inputs)
FieldHub useWhat to write
risk_profile.tierdisplay, governancelow / medium / high.
risk_profile.capabilities_at_riskdisplay, governanceEnum: file_write, cloud_call, payment, publish, delete, private_data_export, external_tool. It never becomes semantic authority identity.
approval_requirements→ authorities columnSame vocabulary; what a human must approve.
approval_scoperuntime policy{grant, ttl_seconds}.
memory_behaviorruntime policy{reads, writes, exports_to_cloud} — required object.
data_accessruntime policy{reads, writes, exports} path classes.
cloud_delegation_policyruntime policynever / ask / allowed_with_grant.
cost_hintsdisplay{model_calls, paid_api}.
entrypointsexecution{canonical_command, agent, terminal}.
benchmark_fixturesqualityPath to .agentlas/routing-benchmarks.jsonl.
locale_coveragedisplay only{primary, ready, partial} for marketplace copy. It must never project into workforce languages.
routing_statusgatedraft → searchable → candidate → routing_ready → trusted. Only a card with real triggers, a real summary and a filled workforce block should claim routing_ready.
routing_status_reasongateWhy it is not ready, when it is not.
card_quality_score, quality, integrity, source, stale, updated_at, agent_card_refpipeline-ownedLeave to lint, publish and the hub. Do not hand-author scores.

Checklist a builder can run

  1. summary names the deliverable in a requester's words, under 240 chars.
  2. 6+ trigger_examples (3 ko / 3 en) and 4+ anti_triggers, all real sentences.
  3. capabilities: 4–8 verb_object ids, checked against community aliases for accidental pulls (contract → legal, and similar).
  4. No description inside required_inputs, optional_inputs, consumes or produces. Explanations go in input_notes / description / known_failure_cases.
  5. workforce.communities, skills, roles, and knowledge use stable open English IDs. Seed aliases are reusable examples, never an allowlist.
  6. risk_profile.capabilities_at_risk lists only real risks.
  7. routing_status is honest.
  8. Validate: the card parses against schemas/routing-card.schema.json.

Verified end to end

templates/routing-card.example.json was compiled through the live hub compiler (compileExactOwnerPrivateWorkforceProfile) on 2026-07-30. Result:

communities  [4] backend-engineering, database-engineering, legal*, software-engineering
skills       [9] api-design, software-architecture, data-modeling + 6 from capabilities
knowledge    [2] api-contract-checklist, database-schema-patterns   (from knowledge/*.md)
inputs       [5] requirement-brief, load-profile, service-context, stack-constraints, text*
outputs      [3] api-contract, data-model, decision-record
runtimes     [4] claude-code, codex, gemini-cli, terminal
languages    [2] en, ko          modalities [1] text
roles        [2] backend-engineer, software-architect
authorities  [1] file-write      forbiddenAuthorities []

The same card written with sentences inside produces/required_inputs produced outputs [6] and inputs [10] full of unmatchable slugs like artifact:architecture-decision-record-naming-the-chosen-component-boundaries-and-…. That is the difference this spec exists to hold.

Two known compiler leaks are visible above and are not the author's fault: legal* (community alias "contract review") and artifact:text* (the type value of an input becoming a concept). Fix them in the compiler, not by contorting the card.

© agentlas-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/routing-card-authoring of agentlas-ai/Agentlas-OS.

Open the folder on GitHubat commit 46420de

Compare with similar skills

Routing Card Authoring next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Routing Card Authoring compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Routing Card Authoring this skillagentlas-ai/Agentlas-OS1.6k—~2.7kAutomated safety check: PassApache-2.0
Agent BuildershareAI-lab/learn-claude-code78k4 repos~1.2kAutomated safety check: PassMIT
Paperclip Create Agentpaperclipai/paperclip100k1 repos~2.1kAutomated safety check: PassMIT
Senior Prompt Engineermaslennikov-ig/claude-code-orchestrator-kit2603 repos~1.4kAutomated safety check: PassCustom licence
Create Agentgnekt/My-Brain-Is-Full-Crew3.9k—~3.1kAutomated safety check: PassCustom licence
Ms Agent Framework RAGshuyu-labs/WebCode278—~1.1kAutomated safety check: PassCustom licence

Similar skills

  • Agent Builder

    shareAI-lab/learn-claude-code

    Design and build AI agents for any domain. An agent skill from shareAI-lab/learn-claude-code.

    78k GitHub starsUsed in 4 repos~1.2k tokens
    AI & LLM EngineeringAuto-check passed

  • Paperclip Create Agent

    paperclipai/paperclip

    Create new agents in Paperclip with governance-aware hiring.

    100k GitHub starsUsed in 1 repo~2.1k tokens
    AI & LLM EngineeringAuto-check passed

  • Senior Prompt Engineer

    maslennikov-ig/claude-code-orchestrator-kit

    Provides reference guides and Python scripts for prompt optimization, RAG evaluation, and agent orchestration when building or tuning LLM systems.

    260 GitHub starsUsed in 3 repos~1.4k tokens
    AI & LLM EngineeringAuto-check passed

  • Create Agent

    gnekt/My-Brain-Is-Full-Crew

    Create a new custom agent from scratch. An agent skill from gnekt/My-Brain-Is-Full-Crew.

    3.9k GitHub stars~3.1k tokensUpdated 3 mo ago
    AI & LLM EngineeringAuto-check passed

  • Ms Agent Framework RAG

    shuyu-labs/WebCode

    Comprehensive guide for building Agentic RAG systems using Microsoft Agent Framework in C.

    278 GitHub stars~1.1k tokensUpdated 3 mo ago
    AI & LLM EngineeringAuto-check passed

  • Google Agents CLI Adk Code

    pifferologo/cloud-agents-cli

    This skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use state management", or needs ADK (Agent…

    129 GitHub starsUsed in 1 repo~768 tokens
    AI & LLM EngineeringAuto-check passed

More from agentlas-ai/Agentlas-OS

All 53 skills in this repo

  • Agentlas Security Scan

    agentlas-ai/Agentlas-OS

    A skill your agent uses when an agent folder must pass the Agentlas Cloud 2-stage security scan (static rules + BYOK LLM judgment) before private sync or public publish, or when asked to…

    1.6k GitHub stars~822 tokensUpdated today
    Auto-check passed

  • Hephaestus Build

    agentlas-ai/Agentlas-OS

    A skill your agent uses when the user types /prompts:hep-build or /agentlas-build, mentions @Hephaestus for build work, asks to create a single Agentlas agent, create a multi-agent team, or package…

    1.6k GitHub stars~2.2k tokensUpdated today
    Auto-check passed

  • Hephaestus Graph

    agentlas-ai/Agentlas-OS

    A skill your agent uses when the user types /hep-graph or asks to create, list, inspect, or request a run of an Agentlas automation graph.

    1.6k GitHub stars~498 tokensUpdated today
    Auto-check passed

  • Hephaestus Upload

    agentlas-ai/Agentlas-OS

    A skill your agent uses when the user types $hephaestus-upload, /hep-upload, or /agentlas-upload, or asks to upload, publish, or list an Agentlas agent or team.

    1.6k GitHub stars~1.3k tokensUpdated today
    Auto-check passed

  • Team Builder Packaging

    agentlas-ai/Agentlas-OS

    A skill your agent uses when generating or auditing a multi-role agent team package with orchestrator, PM Soul, Memory Curator, Policy Gate, workers, eval, QA, handoffs, and runtime adapters.

    1.6k GitHub stars~642 tokensUpdated today
    Auto-check passed

  • Hephaestus Network

    agentlas-ai/Agentlas-OS

    A skill your agent uses when the user types $hephaestus-network, /hep-network, or /agentlas-network, mentions @Hephaestus, or asks Agentlas to staff a durable goal from registered Local, owner…

    1.6k GitHub stars~5.8k tokensUpdated today
    Auto-check passed

Categories

Questions about Routing Card Authoring

What does Routing Card Authoring do?

A skill your agent uses whenever a build emits or repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder, the team builder, and the packager. Routing Card Authoring is an agent skill from agentlas-ai/Agentlas-OS.json — the shared card contract for the single-agent builder, the team builder, and the packager.

When should I use Routing Card Authoring?

Routing Card Authoring fits situations like: repairs .agentlas/routing-card.json — the shared card contract for the single-agent builder; the team builder.

How do I install Routing Card Authoring in Claude Code?

Run `npx skills add agentlas-ai/Agentlas-OS --skill routing-card-authoring -a claude-code`. Or copy the skill folder (skills/routing-card-authoring in agentlas-ai/Agentlas-OS) into .claude/skills/routing-card-authoring in your project. Claude Code loads it when a task matches its description.

How do I install Routing Card Authoring in Codex?

Run `npx skills add agentlas-ai/Agentlas-OS --skill routing-card-authoring -a codex`. Or copy the skill folder (skills/routing-card-authoring in agentlas-ai/Agentlas-OS) into .agents/skills/routing-card-authoring in your project. Codex loads it when a task matches its description.

Can I use Routing Card Authoring in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add agentlas-ai/Agentlas-OS --skill routing-card-authoring -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/routing-card-authoring, .gemini/skills/routing-card-authoring, .github/skills/routing-card-authoring and .opencode/skills/routing-card-authoring in your project.

What does Routing Card Authoring need to run?

SKILL.md names no scripts, command-line tools or credentials: Routing Card Authoring is instructions for the agent only.

Does Routing Card Authoring access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Routing Card Authoring safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Routing Card Authoring use?

Routing Card Authoring is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Routing Card Authoring use?

About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Routing Card Authoring?

Skills that share tags, products or a category with Routing Card Authoring: Agent Builder (shareAI-lab/learn-claude-code, 78k stars), Paperclip Create Agent (paperclipai/paperclip, 100k stars), Senior Prompt Engineer (maslennikov-ig/claude-code-orchestrator-kit, 260 stars) and Create Agent (gnekt/My-Brain-Is-Full-Crew, 3.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Routing Card Authoring?

agentlas-ai (a GitHub organization) maintains it in agentlas-ai/Agentlas-OS, which has 1,576 GitHub stars. The repository holds 53 skills in this directory. The repository was last updated on October 10, 2026.

Source: agentlas-ai/Agentlas-OS on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.