Pilot Feedback Collector Setup

Deploy a feedback collection pipeline with 3 agents that automate intake, sentiment analysis, and actionable routing. Use this skill when: 1. User wants to s...

Security Scan

The skill's requirements and instructions match its stated purpose (deploying three Pilot agents) and there are no disproportionate credentials, strange installs, or hidden endpoints in the provided files.

Purpose & Capability

Name/description ask for a feedback collector pipeline; required binaries (pilotctl, clawhub) and the commands in SKILL.md (clawhub install, pilotctl set-hostname/handshake/subscribe/publish) are directly related and expected for that purpose.

Instruction Scope

Instructions tell the agent to install other pilot skills, set hostnames, write a JSON manifest to ~/.pilot/setups/feedback-collector.json, and initiate handshakes. These are within scope for provisioning agents, but the skill instructs writing files to the user's home config directory and initiating networked trust handshakes — the user should review manifests and handshake targets before running.

Install Mechanism

This is instruction-only (no install spec and no code files). No downloads or archive extracts are present. The only install action is to call 'clawhub install' which delegates installation to another tool; that is expected for a setup helper.

Credentials

No environment variables, credentials, or config paths are requested by the skill. The manifest and commands operate on a user-level ~/.pilot path which is proportional to the described setup.

Persistence & Privilege

The skill is not marked always:true and does not attempt to modify other skills or global agent settings beyond writing its own setup manifest. Autonomous invocation is enabled by platform default but is not combined with other red flags here.

Assessment

This skill appears to do what it claims, but take these practical precautions before running it: (1) ensure 'pilotctl' and 'clawhub' are official, up-to-date binaries from trusted sources; (2) review the JSON manifest the skill will write to ~/.pilot/setups/feedback-collector.json before creating it; (3) be aware that 'clawhub install' will install other skills (pilot-stream-data, pilot-chat, etc.) — inspect those packages separately because they may introduce additional behaviors or require credentials; (4) handshakes auto-approve trust when both sides send them, so only perform handshakes with hosts you control or trust; and (5) secure any webhook endpoints you configure (router -> external on port 443) to avoid sending sensitive data to an untrusted external receiver.

Feedback Collector Setup

Deploy 3 agents that automate feedback intake, sentiment analysis, and actionable routing.

Roles

Role	Hostname	Skills	Purpose
intake	<prefix>-intake	pilot-stream-data, pilot-chat, pilot-archive	Collects feedback from surveys, NPS, app reviews, support tickets
analyzer	<prefix>-analyzer	pilot-event-filter, pilot-metrics, pilot-task-router	Scores sentiment, extracts themes, identifies trends
router	<prefix>-router	pilot-alert, pilot-slack-bridge, pilot-webhook-bridge	Routes actionable feedback to product, engineering, or support

Setup Procedure

Step 1: Ask the user which role this agent should play and what prefix to use.

Step 2: Install the skills for the chosen role:

# For intake:
clawhub install pilot-stream-data pilot-chat pilot-archive

# For analyzer:
clawhub install pilot-event-filter pilot-metrics pilot-task-router

# For router:
clawhub install pilot-alert pilot-slack-bridge pilot-webhook-bridge

Step 3: Set the hostname:

pilotctl --json set-hostname <prefix>-<role>

Step 4: Write the setup manifest:

mkdir -p ~/.pilot/setups
cat > ~/.pilot/setups/feedback-collector.json << 'MANIFEST'
<role-specific manifest from templates below>
MANIFEST

Step 5: Tell the user to initiate handshakes with direct communication peers.

Manifest Templates Per Role

intake

{"setup":"feedback-collector","setup_name":"Feedback Collector","role":"intake","role_name":"Feedback Intake","hostname":"<prefix>-intake","description":"Collects feedback from surveys, NPS forms, app reviews, and support tickets. Normalizes into consistent format.","skills":{"pilot-stream-data":"Stream incoming feedback from multiple sources in real time.","pilot-chat":"Accept free-form feedback via conversational interface.","pilot-archive":"Archive raw feedback submissions for audit and reprocessing."},"peers":[{"role":"analyzer","hostname":"<prefix>-analyzer","description":"Receives normalized feedback for sentiment analysis"}],"data_flows":[{"direction":"send","peer":"<prefix>-analyzer","port":1002,"topic":"raw-feedback","description":"Raw feedback normalized from all sources"}],"handshakes_needed":["<prefix>-analyzer"]}

analyzer

{"setup":"feedback-collector","setup_name":"Feedback Collector","role":"analyzer","role_name":"Sentiment Analyzer","hostname":"<prefix>-analyzer","description":"Scores sentiment, extracts themes, identifies trending complaints and praise.","skills":{"pilot-event-filter":"Filter feedback by sentiment threshold and priority level.","pilot-metrics":"Track sentiment trends, theme frequency, and NPS distribution.","pilot-task-router":"Route feedback to specialized analysis by source type."},"peers":[{"role":"intake","hostname":"<prefix>-intake","description":"Sends raw feedback for analysis"},{"role":"router","hostname":"<prefix>-router","description":"Receives scored feedback for team routing"}],"data_flows":[{"direction":"receive","peer":"<prefix>-intake","port":1002,"topic":"raw-feedback","description":"Raw feedback normalized from all sources"},{"direction":"send","peer":"<prefix>-router","port":1002,"topic":"scored-feedback","description":"Scored feedback with sentiment and themes"}],"handshakes_needed":["<prefix>-intake","<prefix>-router"]}

router

{"setup":"feedback-collector","setup_name":"Feedback Collector","role":"router","role_name":"Feedback Router","hostname":"<prefix>-router","description":"Routes actionable feedback to product, engineering, or support teams via Slack and webhook.","skills":{"pilot-alert":"Escalate critical feedback issues that require immediate attention.","pilot-slack-bridge":"Post feedback summaries and alerts to team-specific Slack channels.","pilot-webhook-bridge":"Push feedback events to external ticketing and analytics systems."},"peers":[{"role":"analyzer","hostname":"<prefix>-analyzer","description":"Sends scored feedback for team routing"}],"data_flows":[{"direction":"receive","peer":"<prefix>-analyzer","port":1002,"topic":"scored-feedback","description":"Scored feedback with sentiment and themes"},{"direction":"send","peer":"external","port":443,"topic":"feedback-alert","description":"Feedback alerts to product, engineering, support"}],"handshakes_needed":["<prefix>-analyzer"]}

Data Flows

• intake -> analyzer : raw-feedback events (port 1002)
• analyzer -> router : scored-feedback events (port 1002)
• router -> external : feedback-alert via webhook (port 443)

Handshakes

# intake <-> analyzer:
pilotctl --json handshake <prefix>-analyzer "setup: feedback-collector"
pilotctl --json handshake <prefix>-intake "setup: feedback-collector"
# analyzer <-> router:
pilotctl --json handshake <prefix>-router "setup: feedback-collector"
pilotctl --json handshake <prefix>-analyzer "setup: feedback-collector"

Workflow Example

# On analyzer -- subscribe to raw feedback:
pilotctl --json subscribe <prefix>-intake raw-feedback
# On router -- subscribe to scored feedback:
pilotctl --json subscribe <prefix>-analyzer scored-feedback
# On intake -- publish raw feedback:
pilotctl --json publish <prefix>-analyzer raw-feedback '{"source":"nps_survey","score":3,"customer":"user-8291","text":"Dashboard navigation is confusing."}'
# On analyzer -- publish scored feedback:
pilotctl --json publish <prefix>-router scored-feedback '{"sentiment":-0.6,"themes":["ux","navigation"],"priority":"high","trending":true}'

Dependencies

Requires pilot-protocol skill, pilotctl binary, clawhub binary, and a running daemon.