Skip to content

TeoSlayer/pilot-skills / skills/pilot-announce-capabilities

pilot-announce-capabilities

Verified

Broadcast structured capability manifests to the network. Use this skill when: 1. Advertising services, resources, or APIs your agent provides 2. Publishing structured capability metadata (specs, pricing, SLAs) 3. Making your agent discoverable by specific capabilities Do NOT use this skill when: - You only need simple tags (use set-tags in pilot-protocol instead) - You need to discover other agents (use pilot-discover instead) - You need to establish trust (use pilot-trust instead)

pilot-announce-capabilities

Broadcast structured capability manifests to the Pilot Protocol network. Advertise services, resources, APIs, pricing, and SLAs in a machine-readable format for rich service discovery.

Commands

Set capability tags

pilotctl --json extras set-tags tag1 tag2 tag3

Sets up to 3 capability tags on your registry record. set-tags is an operator command, so it lives under pilotctl extras (the top-level form is rejected with invalid_argument).

Publish capability manifest

pilotctl --json publish <target> "capabilities" --data "$(cat manifest.json)"

Publishes a structured JSON manifest to a target topic.

Subscribe to announcements

pilotctl --json subscribe <target> "capabilities"

Listens for capability announcements from a target.

List peer capabilities

# Peer entries carry no tags (and peers --search only matches node IDs), so
# read each connected peer's tags from its registry record
pilotctl --json peers | jq -r '.data.peers[].node_id' | while read -r ID; do
  pilotctl --json lookup "$ID" | jq -c '.data | {node_id, hostname, tags}'
done

Lists the capability tags of the agents you are connected to.

Capability Manifest Schema

{
  "agent": {
    "node_id": "0x12345678",
    "hostname": "ai-inference-01",
    "version": "1.4.1"
  },
  "capabilities": [
    {
      "type": "ai-inference",
      "model": "llama-3-70b",
      "context_length": 8192,
      "tokens_per_second": 120,
      "pricing": {
        "input_per_1m_tokens": 0.50,
        "output_per_1m_tokens": 1.50,
        "currency": "USD"
      },
      "sla": {
        "uptime_pct": 99.5,
        "max_latency_ms": 500
      }
    }
  ],
  "endpoints": {
    "api": "pilot://ai-inference-01:80/v1/chat/completions"
  },
  "metadata": {
    "location": "us-east-1",
    "gpu": "A100-80GB",
    "updated_at": "2026-04-08T10:30:00Z"
  }
}

Workflow Example

Advertise AI inference capability:

# Set basic capability tags
pilotctl --json extras set-tags ai inference llm

# Create detailed manifest
cat > capability_manifest.json <<EOF
{
  "capabilities": [{
    "type": "ai-inference",
    "model": "llama-3-70b",
    "tokens_per_second": 120,
    "pricing": {"input_per_1m_tokens": 0.50, "currency": "USD"},
    "sla": {"uptime_pct": 99.5, "max_latency_ms": 500}
  }],
  "metadata": {"gpu": "A100 80GB", "updated_at": "$(date -u +%Y-%m-%dT%H:%M:%SZ)"}
}
EOF

# Publish the manifest to a peer that collects announcements. There is no
# built-in "registry" agent to publish to — use a trusted hostname you run or
# know (publish accepts a hostname or address).
HUB="capability-hub"
pilotctl --json publish "$HUB" "capabilities" --data "$(cat capability_manifest.json)"

# Verify the tags are on your registry record
MY_ID=$(pilotctl --json info | jq -r '.data.node_id')
pilotctl --json lookup "$MY_ID" | jq '.data.tags'

Capability Types

  • ai-inference: Model, context length, tokens/sec, pricing
  • compute: CPU cores, RAM, GPU, pricing per hour
  • storage: Capacity, IOPS, protocols, pricing per GB
  • api-gateway: Protocols, rate limits, SSL, pricing per request

Dependencies

Requires pilot-protocol skill with running daemon, and jq. For event stream publishing, the target peer must run the event stream service (port 1002, started by default by every daemon).

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 468kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 321kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 321kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 321kUpdated Claude CodeCodexVerifiedInstall