An evaluation-driven workflow for improving agent prompts while preserving their task contract and instruction ownership.
Editorial review checked 2026-10-29 against this instruction snapshot. Supporting scripts and runtime behavior remain outside this editorial review.
This source separates new prompts, refinement, provider migration and failure diagnosis. It inventories context and constraints, then proposes a baseline and a small set of candidate changes. Stable repository references are kept explicit rather than copied into a growing prompt.
Teams with recurring prompt failures who can supply representative inputs, expected outputs and a way to compare alternatives.
Prompt wording may not be the actual bottleneck: retrieval, model choice, missing tools or an ambiguous task can dominate. Model-family notes are version-sensitive. Running external evaluations can incur costs even though reading this skill does not.
Unknown; inspect source requirements and supporting scripts.
Inspect the pinned source snapshot and its supporting files first. The command selects this skill through the open-source skills CLI. Node.js and agent-specific setup may be required. Installation can download code. Catalog ingestion never executes skill instructions or install scripts.
npx skills add https://github.com/getsentry/skills/tree/d12f250460ffd961f549dbebf7eeb0cf4afd4c2e/skills/prompt-optimizer --skill prompt-optimizerReview pinned skill folder ↗ · CLI documentation ↗
After source review, use npx skills check and npx skills update to inspect and apply upstream updates. Re-check changed instructions before running them.
Static review only. This is not a security audit or a guarantee of safety.
SKILL.md text only; scripts, dependencies and runtime behavior not audited.
No configured text patterns matched. Unmatched or concealed behavior can still exist.
Allowed tools declaration: Unknown. Runtime permissions and network behavior: Unknown.
86/100 · Metadata coverage 100% · Model 1.0
Scores summarize observable discovery signals. Popularity and repository activity apply to the whole repository. Unknown runtime safety earns no certification.
Record the required behavior and failure cases before rewriting. Evaluate a baseline, identify one recurring cause and compare a few targeted candidates on the same examples.
Check held-out cases as well as the examples used for revision. Track constraint compliance, quality, latency and cost separately, and retain the baseline when a cleaner-looking prompt has not demonstrated an improvement.
Related by category; compare scope and dependencies before choosing an alternative.
Explore diagnosing superpowers as a automation & agents workflow from obra/superpowers. Compare its source, maintenance and review signals before choosing it for your agent.
Explore using superpowers as a automation & agents workflow from obra/superpowers. Compare its source, maintenance and review signals before choosing it for your agent.
Explore 10 andruia skill smith as a automation & agents workflow from sickn33/agentic-awesome-skills. Compare its source, maintenance and review signals before choosing it for your agent.
A concise repository-instruction workflow that derives agent guidance from files and commands that actually exist.
Explore blog writing guide as a writing & documents workflow from getsentry/skills. Compare its source, maintenance and review signals before choosing it for your agent.
Explore brand guidelines as a writing & documents workflow from getsentry/skills. Compare its source, maintenance and review signals before choosing it for your agent.