Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Tracks the cost-to-value ratio of agent sessions by logging token usage alongside a subjective value category. Stores sessions locally and surfaces patterns like high-cost/low-value work or zero-value heartbeat cycles.
Most cost dashboards show token spend without tying it to outcomes; this links cost directly to delivered value in a single log.
When to use it
- Logging value delivered after fixing a CI pipeline with an agent
- Identifying task types that consistently produce zero output
- Weekly review of which sessions were worth their token cost
- Deciding whether to start a task based on past ROI patterns
- Batching heartbeat checks after noticing they rarely produce verifiable artifacts
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: log a session where I fixed the CI pipeline
- 1Run ./track.sh quick with the task description
- 2Assign a value category (high, medium, low, or zero)
- 3Provide the estimated token count for the session
- 4Session is appended to ~/.clawdbot/session-costs.json
Session logged; ./track.sh stats shows the new entry and updated cost-to-value summary