Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add marcusgoll/Spec-Flow --skill optimization-phasegit clone --depth 1 https://github.com/marcusgoll/Spec-FlowWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/marcusgoll/spec-flow/optimization-phase)<a href="https://agentmods.dev/skills/marcusgoll/spec-flow/optimization-phase"><img src="https://agentmods.dev/badge/skills/marcusgoll/spec-flow/optimization-phase/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/marcusgoll/spec-flow/optimization-phase"><img src="https://agentmods.dev/badge/skills/marcusgoll/spec-flow/optimization-phase.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00042 | $0.04247 |
| Opus 5.5 | $0.00017 | $0.01699 |
| Sonnet 5.5 | $0.00008 | $0.00849 |
| Haiku 4.5 | $0.00004 | $0.00425 |
Grade A, and why
optimization-phase scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -w "%{time_total}\n" -o /dev/null -s "http://localhost:3000/api/endpoint" How it starts
The opening of the file, as written. The whole thing — 642 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This skill orchestrates the /optimize phase, which runs after /implement and before /preview in the deployment workflow.
Inputs: Implemented code (from /implement), spec.md (success criteria) Outputs: optimization-report.md, code-review-report.md Expected duration: 1-2 hours
<quick_start> Execute optimization quality gates systematically:
- Performance Benchmarking - API p50 <200ms, p95 <500ms, page load <2s
- Accessibility Audit - WCAG 2.1 AA compliance (if UI feature)
- Security Review - npm audit, dependency scanning, OWASP top 10
- Code Quality Review - DRY violations <3, test coverage ≥80%
- Generate Reports - optimization-report.md + code-review-report.md
Key principle: All blocking quality gates must pass before deployment. </quick_start>
If prerequisites not met, return to /implement phase.
Validate API response times and page load performance against spec.md targets.
API Performance Testing:
# Measure API endpoint response times (10 requests)
for i in {1..10}; do
curl -w "%{time_total}\n" -o /dev/null -s "http://localhost:3000/api/endpoint"
done | sort -n
# Calculate p50 (median): 5th value
# Calculate p95 (95th percentile): 10th value
Page Load Testing:
# Run Lighthouse CI for page load metrics
npm install -g @lhci/cli
lhci autorun --url=http://localhost:3000
# Check metrics:
# - First Contentful Paint (FCP): <1.8s
# - Largest Contentful Paint (LCP): <2.5s
# - Time to Interactive (TTI): <3.8s
Targets from spec.md:
- API p50 (median): <200ms
- API p95 (95th percentile): <500ms
- Page load (initial): <2 seconds
- Time to Interactive (TTI): <3 seconds
Blocking: If performance targets not met, optimization required before deployment.
See resources/performance-benchmarking.md for detailed testing procedures.
Validate WCAG 2.1 AA compliance for all UI changes.
Lighthouse CI Audit:
# Run Lighthouse accessibility audit
lhci autorun --collect.settings.onlyCategories=accessibility
# Target score: ≥95/100
axe-core Validation:
# Install axe-core DevTools extension
# Or use axe CLI:
npm install -g @axe-core/cli
axe http://localhost:3000 --tags wcag2aa
WCAG 2.1 AA Requirements:
- Color contrast ≥4.5:1 (text), ≥3:1 (UI components)
- Keyboard navigation (all interactive elements accessible)
- Screen reader compatibility (ARIA labels, semantic HTML)
- Focus indicators visible (outline, border, background change)
Blocking (if UI feature): All WCAG 2.1 AA violations must be fixed.
Skip if: No UI changes in this feature (backend/API only).
See resources/accessibility-audit.md for comprehensive checklist.
Scan for vulnerabilities in dependencies and code.
Dependency Vulnerability Scan:
# npm audit for Node.js projects
npm audit --audit-level=moderate
# Check for critical and high severity vulnerabilities
# Fix: npm audit fix (or manual upgrade)
OWASP Top 10 Checks:
- SQL Injection: Parameterized queries, no string concatenation
- XSS: Input sanitization, output encoding
- CSRF: CSRF tokens on state-changing endpoints
- Authentication: Secure password hashing (bcrypt), JWT validation
- Sensitive Data Exposure: No secrets in code/logs, HTTPS enforced
Secret Scanning:
# Check for accidentally committed secrets
git secrets --scan
# Or use gitleaks:
gitleaks detect --source . --verbose
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- examples.md 247 B
- reference.md 7.4 KB
- resources/accessibility-audit.md 1.7 KB
- resources/code-quality-review.md 883 B
- resources/code-review-checklist.md 997 B
- resources/common-mistakes.md 1.3 KB
- resources/performance-benchmarking.md 1.4 KB
- resources/report-generation.md 1.6 KB
- resources/security-review.md 1.2 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 642 lines · 42 tokens per session scan A c07d926f1b15
optimization-phase is a skill published in the GitHub repository marcusgoll/Spec-Flow (96 stars, last pushed 23d ago), licensed MIT. It adds 42 tokens to every session and 4,247 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
agent-orchestration-improve-agent
Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.
android-ui-journey-testing
XML-specified Android UI journey testing, interactive step execution, assertion verification, and JSON outcome reporting.
quinn
Verifies specified behavior through appropriate tests and truthful coverage reports.
agent-qa-result-triage
Triage failed Agent QA runs with MCP evidence, artifacts, logs, fixed failure categories, confidence, and actionable next steps.
android_ui_verification
Automated end-to-end UI testing and verification on an Android Emulator using ADB.
team-qa
Orchestrate the QA team through a full testing cycle — qa-lead strategy and test plan, qa-tester case writing, execution, sign-off.