Performance Goal
This is an OMH performance-goal workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
Why This Exists
performance-goal exists to keep optimization work explicit, evidence-backed, and inside the Hermes/executor boundary instead of relying on ad hoc chat narration.
Do Not Use When
- The ask is to find where performance problems are, or to fix multiple unscoped hotspots across domains; use
ultraperf.
Examples
Good example:
- Prompt: performance-goal: benchmark recommendation latency, optimize hot paths safely, and prove no regressions.
- Expected behavior: Create a measurement-led optimization loop with baseline, change, verification, and regression evidence.
- Why: The request is performance optimization and needs measured before/after proof.
Bad example:
- Prompt: performance-goal: treat casual chat or unaccepted work as if this workflow already produced verified results.
- Expected behavior: Ask a clarification question or route to a narrower workflow instead of forcing
performance-goal.
- Why: The request lacks the required inputs or would overclaim work that Hermes did not observe.
Completion Checklist
- Confirm the workflow target, evidence boundary, and stop condition are named.
- Report which outputs are prepared, observed, blocked, or missing.
- Name the smallest next verification or handoff instead of claiming completion from narration.
Recovery Notes
- If required context is missing, ask one blocking question or route back to the narrower workflow.
- If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
Use When
Use when the goal is measurable performance improvement with evaluator evidence.
Strong routing signals: `performance-goal`, `performance goal`, `latency`, `throughput`, `benchmark`
Catalog Metadata
Category: optimization
Phase: measurement
Quality tier: measurement-gated
Reasoning demand: heavy
Quality bar:
- Name the metric, baseline, budget, and benchmark command before optimizing.
- Treat code-level optimization as executor work when edits are required.
- Report deltas only from observed benchmark evidence.
Required inputs:
- metric
- baseline
- budget
- benchmark command
Expected outputs:
- measurement delta
- implementation summary
- benchmark evidence
Artifact expectations:
- baseline and final benchmark evidence
Safety rules:
- Do not imply hidden Hermes runtime behavior.
- Use the smallest verification that can prove the claim.
Runtime Evidence
Use the current host's own tools and subagent/task mechanism when available;
otherwise run the same lanes sequentially or name the unavailable capability.
A prepared plan, handoff, checklist, or skill installation is not execution,
review, CI, merge-readiness, or merge evidence. Report actual tool results or
not_observed / not_available; never invent dispatch or host accounting.
Treat supplied context as advisory, not proof of hidden memory reads or writes.
State scope, constraints, verification, and the stop condition before work.
Supporting paths are relative to this skill directory; sibling skill paths are
relative to its parent. Resolve them from the host-provided skill base directory
({baseDir} on hosts that provide it), never a hardcoded install location.
A named workflow not installed here is unavailable, not permission to emulate
its host-specific capabilities. Verify through the real surface before done.
1---2name: omh-performance-goal-23description: "[omh] Hermes adaptation for measurable performance-goal execution. Use when the user says: performance-goal, performance goal, latency, throughput, benchmark."4metadata:5 source: https://skillmd.com/skills/rlaope/omh-performance-goal-26 hermes:7 tags: [workflow, oh-my-hermes, optimization]8 category: optimization9 phase: measurement10 role: tracker11 quality_tier: measurement-gated12---1314# Performance Goal1516This is an OMH `performance-goal` workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).1718## Why This Exists1920`performance-goal` exists to keep `optimization` work explicit, evidence-backed, and inside the Hermes/executor boundary instead of relying on ad hoc chat narration.2122## Do Not Use When2324- The ask is to find where performance problems are, or to fix multiple unscoped hotspots across domains; use `ultraperf`.2526## Examples2728Good example:2930- Prompt: performance-goal: benchmark recommendation latency, optimize hot paths safely, and prove no regressions.31- Expected behavior: Create a measurement-led optimization loop with baseline, change, verification, and regression evidence.32- Why: The request is performance optimization and needs measured before/after proof.3334Bad example:3536- Prompt: performance-goal: treat casual chat or unaccepted work as if this workflow already produced verified results.37- Expected behavior: Ask a clarification question or route to a narrower workflow instead of forcing `performance-goal`.38- Why: The request lacks the required inputs or would overclaim work that Hermes did not observe.3940## Completion Checklist4142- Confirm the workflow target, evidence boundary, and stop condition are named.43- Report which outputs are prepared, observed, blocked, or missing.44- Name the smallest next verification or handoff instead of claiming completion from narration.4546## Recovery Notes4748- If required context is missing, ask one blocking question or route back to the narrower workflow.49- If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.50515253## Use When5455Use when the goal is measurable performance improvement with evaluator evidence.5657 Strong routing signals: `performance-goal`, `performance goal`, `latency`, `throughput`, `benchmark`5859## Catalog Metadata6061Category: `optimization`62Phase: `measurement`63Quality tier: `measurement-gated`64Reasoning demand: `heavy`6566Quality bar:6768- Name the metric, baseline, budget, and benchmark command before optimizing.69- Treat code-level optimization as executor work when edits are required.70- Report deltas only from observed benchmark evidence.7172Required inputs:7374- metric75- baseline76- budget77- benchmark command7879Expected outputs:8081- measurement delta82- implementation summary83- benchmark evidence8485Artifact expectations:8687- baseline and final benchmark evidence8889Safety rules:9091- Do not imply hidden Hermes runtime behavior.92- Use the smallest verification that can prove the claim.9394## Runtime Evidence9596Use the current host's own tools and subagent/task mechanism when available;97otherwise run the same lanes sequentially or name the unavailable capability.98A prepared plan, handoff, checklist, or skill installation is not execution,99review, CI, merge-readiness, or merge evidence. Report actual tool results or100`not_observed` / `not_available`; never invent dispatch or host accounting.101Treat supplied context as advisory, not proof of hidden memory reads or writes.102State scope, constraints, verification, and the stop condition before work.103Supporting paths are relative to this skill directory; sibling skill paths are104relative to its parent. Resolve them from the host-provided skill base directory105(`{baseDir}` on hosts that provide it), never a hardcoded install location.106A named workflow not installed here is unavailable, not permission to emulate107its host-specific capabilities. Verify through the real surface before done.