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 agentmods add skills/marcusgoll/spec-flow/analysis-phasenpx skills add marcusgoll/Spec-Flow --skill analysis-phasegit clone --depth 1 https://github.com/marcusgoll/Spec-FlowWhat 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 | $0.00049 | $0.00977 |
| Opus 5 | $0.00024 | $0.00489 |
| Sonnet 5 | $0.00010 | $0.00195 |
| Haiku 4.5 | $0.00005 | $0.00098 |
Grade A, and why
analysis-phase scanned grade A with 0 findings 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<quick_start> Validate cross-artifact consistency and detect breaking changes:
- Check spec → plan → tasks alignment
- Detect breaking changes (API, database, UI, auth)
- Validate dependencies
- Generate analysis-report.md with findings
Inputs: spec.md, plan.md, tasks.md Outputs: analysis-report.md </quick_start>
Read spec.md requirements, verify each has corresponding plan component. Flag missing mappings in analysis report.
See references/examples.md for grep commands.
Read plan.md components, verify each broken into tasks in tasks.md.
Validate:
- Each plan component has ≥1 task
- Task acceptance criteria match spec success criteria
- No orphaned tasks (tasks without plan component)
Scan for patterns indicating breaking changes:
- API changes: Endpoint signature modifications, required parameter additions, response format changes
- Database changes: Schema modifications, required field additions, migrations affecting existing data
- UI changes: Component interface changes, prop requirement additions
- Auth changes: Permission model modifications, authentication flow changes
Flag with impact level (Low/Medium/High) using reference rubric.
Cross-reference:
- Imports and integrations mentioned in plan
- External dependencies in tasks
- Integration points in spec
Verify all dependencies documented and accounted for.
Create specs/NNN-slug/analysis-report.md:
# Analysis Report
## Consistency Check
- Spec-Plan: [✓/✗] Description
- Plan-Tasks: [✓/✗] Description
## Breaking Changes
- [High/Medium/Low] Description and impact
## Dependency Validation
- [✓/✗] Dependencies documented
## Recommendations
- Action items to fix inconsistencies
Update state.yaml: analysis.status = completed
- All spec requirements have plan component coverage
- All plan components have task breakdown
- Breaking changes flagged with impact level (Low/Medium/High)
- Analysis report generated with actionable findings
- No orphaned artifacts (tasks without plan, plan without spec)
<anti_patterns> ❌ Don't: Assume no breaking changes without scanning ✅ Do: Explicitly check API signatures, database schema, required fields, auth changes
Why: Breaking changes missed in analysis cause production issues post-deployment
Why: Missing mappings lead to incomplete implementation
Why: Undocumented dependencies cause integration failures
Why: Actionable findings enable quick remediation </anti_patterns>
<success_criteria>
- All spec requirements mapped to plan components (no gaps)
- All plan components broken into tasks (no orphans)
- Breaking changes documented with impact level (High/Medium/Low)
- Dependencies validated and documented
- analysis-report.md generated with actionable findings
- state.yaml updated (analysis.status = completed) </success_criteria>
What ships with it
4 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.
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.
- 2d ago First seen · 156 lines · 49 tokens per session scan A 5cfcd4d2ff4b
analysis-phase is a skill published in the GitHub repository marcusgoll/Spec-Flow (91 stars, last pushed 4mo ago), licensed MIT. It adds 49 tokens to every session and 977 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
writing-for-agents
Writing documents for agents. Use when creating or editing a skill, an AGENTS.md or CLAUDE.md, a subagent definition, a prompt file, or any doc an agent reaches through a pointer.
tdd
Fix a bug test-first: a focused regression test that fails before the fix and passes after it.
grilling
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
advogado-criminal
Advogado criminalista especializado em Maria da Penha, violencia domestica, feminicidio, direito penal brasileiro, medidas protetivas, inquerito policial e acao penal.
agent-evaluation-reporting
Use when summarizing agent evaluations where autonomous, assisted, failed, timed-out, or invalid outcomes must remain distinct and comparable.
agent-orchestration-multi-agent-optimize
Optimize multi-agent systems with coordinated profiling, workload distribution, and cost-aware orchestration. Use when improving agent performance, throughput, or reliability.