MiaoDX/roboclaws / skills/report-performance-analysis
report-performance-analysis
VerifiedUse this skill whenever the user asks whether a Roboclaws live-agent report run is faster, wants to compare Agent SDK/Codex/Claude cleanup reports, asks about model-call work, residual latency, speedup validity, or report performance under provider/network variance. The skill extracts sanitized JSON metrics first and answers only from generated metric/comparison artifacts, not from report.html scraping or raw prompts.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.4 KB |
| references/metric-contract.md | markdown | 1.8 KB |
| skill.json | config | 1.8 KB |
| templates/comparison-manifest.json | config | 340 B |
references/metric-contract.md
View on GitHub# Report Performance Metric Contract
The source of truth is `roboclaws_report_performance_metrics_v1`, produced by
`roboclaws.reports.live_performance`.
## Packet Sections
- `run_identity`: surface, intent, task name, agent engine, provider profile,
model, evidence lane, seed, and profile id.
- `quality`: checker state, terminal state, cleanup/completion status, restored
counts, restoration and sweep rates, disturbances, semantic accepted count,
and failed/no-op tool count.
- `call_counts`: model calls, agent attempts, continuations, MCP tool calls,
per-tool counts, and non-tool turns when available.
- `model_work`: input, cached input, uncached input, output, reasoning, image
counts/pixels, percentiles, and explicit unavailable metrics.
- `timing`: observed wall time, runner agent time, MCP time, between-tool gap,
tool handler time, robot-view capture time, observed model API time,
estimated model work, residual latency, and non-model time.
## Per-Call Rows
`model_call_metrics.jsonl` uses `roboclaws_model_call_metric_v1`.
Allowed row fields are sanitized counts, timing, source, status, failure class,
and limitations. Do not add raw prompts, model text, function inputs/outputs,
full tool payload bodies, credentials, private evaluator truth, or compact
continuation packets.
## Speed Claim Boundary
A single run is diagnostic only. A speedup claim needs:
- explicit baseline or comparison manifest;
- same-or-better quality, or a recorded waiver;
- comparable run identity, or a diagnostic apples-to-oranges label;
- repeated rows or an explicit repeatability waiver for publishable claims;
- calibration metadata before treating normalized model time as authoritative.
When calibration coefficients are unavailable, report
`estimated_model_work_s.available=false` and keep residual latency diagnostic.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 472k | Updated | Claude CodeCodex | Verified | Install |
| 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: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| 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 available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |