ai-feature-evaluation
Evaluate AI and agent features for grounded quality, safety, tool authority, and fallback.
Skill by Sertxito
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/sertxito-ai-feature-evaluation
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: ml, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/ai-feature-evaluation
What's inside
- Skill name
ai-feature-evaluation- Description
- Evaluate AI and agent features for grounded quality, safety, tool authority, and fallback. Use for LLM prompts, retrieval, model changes, or agent workflows.
- License
- MIT
- Instructions
- 14 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# AI feature evaluation ## Procedure 1. Define intended behavior, prohibited behavior, and measurable acceptance criteria. 2. Inspect model, prompt, retrieval sources, data flow, and tool permissions. 3. Build representative, privacy-safe evaluation cases including adversarial cases. 4. Measure quality, grounding, refusal, latency, cost, and fallback behavior. 5. Compare with the approved baseline and report regressions and uncertainty. ## Output Evaluation version, dataset provenance, results, failure examples, thresholds, limitations, and release recommendation. Never claim model output is deterministic.
From Sertxito/mcp-efficiency-engine, MIT.
More from Sertxito/mcp-efficiency-engine
- 0.561score
ai-platform-audit
Sertxito
Review AI, retrieval, agent, and MCP systems for grounding, evaluation, permissions, privacy, observability, and cost evidence.
SkillAFreshness: AClaude Code +9 - 0.561score
api-contract-review
Sertxito
Review API and event contracts for compatibility, validation, security, and failure behavior.
SkillAFreshness: AClaude Code +9 - 0.561score
application-insights-instrumentation
Sertxito
Instrument Azure applications with Application Insights while preserving privacy, sampling, correlation, and operational signal quality.
SkillAFreshness: AClaude Code +9 - 0.561score
backend-change
Sertxito
Implement and verify a focused backend or service change.
SkillAFreshness: AClaude Code +9 - 0.561score
backend-mcp-package-authoring
Sertxito
Build maintainable .NET or Python MCP servers and publishable Python packages with explicit protocol, packaging, and validation contracts.
SkillAFreshness: AClaude Code +9 - 0.561score
ci-cd-review
Sertxito
Review CI/CD pipelines for reproducibility, permissions, artifact integrity, and promotion safety.
SkillAFreshness: AClaude Code +9 - 0.561score
cloud-release-cost-preflight
Sertxito
Check cloud deployment prerequisites, least-privilege role needs, quota, and current pricing before proposing a release.
SkillAFreshness: AClaude Code +9 - 0.561score
cloud-review
Sertxito
Review a declared cloud environment for identity, networking, reliability, recovery, observability, and cost using provider-specific evidence.
SkillAFreshness: AClaude Code +9 - 0.561score
data-pipeline-quality
Sertxito
Design or review data pipeline correctness, lineage, quality, and replay behavior.
SkillAFreshness: AClaude Code +9 - 0.561score
database-change-operations
Sertxito
Assess operational risk for database changes, query tuning, and data operations.
SkillAFreshness: AClaude Code +9 - 0.559score
cloud-architecture-planning
Sertxito
Plan or review cloud architecture, scalability, cost, capacity, identity, and deployment options.
SkillAFreshness: AClaude Code +9 - 0.558score
database-operations-reliability
Sertxito
Assess database jobs, high availability, recovery, and monitoring baselines without changing live systems.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 19, 2026 (11 days ago)
- Last release
- Oct 22, 2026 (7 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Sertxito
- 0.561score
ai-platform-audit
Sertxito
Review AI, retrieval, agent, and MCP systems for grounding, evaluation, permissions, privacy, observability, and cost evidence.
SkillAFreshness: AClaude Code +9 - 0.561score
application-insights-instrumentation
Sertxito
Instrument Azure applications with Application Insights while preserving privacy, sampling, correlation, and operational signal quality.
SkillAFreshness: AClaude Code +9 - 0.561score
api-contract-review
Sertxito
Review API and event contracts for compatibility, validation, security, and failure behavior.
SkillAFreshness: AClaude Code +9 - 0.561score
backend-change
Sertxito
Implement and verify a focused backend or service change.
SkillAFreshness: AClaude Code +9 - 0.561score
backend-mcp-package-authoring
Sertxito
Build maintainable .NET or Python MCP servers and publishable Python packages with explicit protocol, packaging, and validation contracts.
SkillAFreshness: AClaude Code +9 - 0.561score
ci-cd-review
Sertxito
Review CI/CD pipelines for reproducibility, permissions, artifact integrity, and promotion safety.
SkillAFreshness: AClaude Code +9
Cite this page
ai-feature-evaluation skill. Agent Mart. https://agentmart.dev/skills/sertxito-ai-feature-evaluation (last updated 2026-10-24)