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 commands/madappgang/claude-code/featuregit clone --depth 1 https://github.com/MadAppGang/claude-codeWrote 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/commands/madappgang/claude-code/feature)<a href="https://agentmods.dev/commands/madappgang/claude-code/feature"><img src="https://agentmods.dev/badge/commands/madappgang/claude-code/feature.svg" alt="Measured on agentmods" 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 | $0.00010 | $0.14921 |
| Opus 5 | $0.00005 | $0.07461 |
| Sonnet 5 | $0.00002 | $0.02984 |
| Haiku 4.5 | $0.00001 | $0.01492 |
Grade A, and why
feature 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 yesterday.
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.
if curl -s ${test_url} > /dev/null 2>&1; then How it starts
The opening of the file, as written. The whole thing — 1,911 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Key principle: NEVER claim completion without real evidence (screenshots,
browser navigation, actual user flow verification).
<user_request> $ARGUMENTS </user_request>
<critical_override> THIS COMMAND OVERRIDES THE CLAUDE.md TASK ROUTING TABLE FOR AGENT SELECTION.
WHY: This 8-phase workflow uses DIFFERENT agents for each phase. The CLAUDE.md routing table would incorrectly substitute agents (e.g., dev:architect for implementation, code-analysis:detective for debugging). Each phase MUST use its designated agent.
AGENT RULES FOR THIS COMMAND:
- Stack detection → dev:stack-detector agent (subagent_type: "dev:stack-detector")
- Architecture/planning → dev:architect agent (subagent_type: "dev:architect")
- Plan review (external models via claudish) → dev:architect agent
- Implementation → dev:developer agent (subagent_type: "dev:developer")
- Code review (external models via claudish) → agentdev:reviewer or dev:architect agent
- Test creation → dev:test-architect agent (subagent_type: "dev:test-architect")
- Real validation → Orchestrator (Chrome MCP tools directly)
DO NOT substitute agents across phases. Each phase has specific agent requirements. DO NOT use code-analysis:detective for any phase (READ-ONLY, cannot write code). DO NOT use dev:researcher for any phase (research only, not in this workflow). </critical_override>
Before starting, create comprehensive todo list:
1. PHASE 0: Session initialization
2. PHASE 1: Requirements + validation setup + iteration config
3. PHASE 2: Research (optional)
4. OUTER LOOP (Phases 3-7):
- PHASE 3: Multi-model planning
- PHASE 4: Implementation
- PHASE 5: Code review loop
- PHASE 6: Black box unit testing
- PHASE 7: Real 3rd party validation
5. PHASE 8: Completion (only after Phase 7 passes)
Update continuously as you progress.
Mark only ONE task as in_progress at a time.
</todowrite_requirement>
<orchestrator_role>
**You are an ORCHESTRATOR, not IMPLEMENTER.**
**You MUST:**
- Use Task tool to delegate ALL work to agents
- Use Tasks to track full lifecycle
- Enforce quality gates between phases
- Respect iteration limits (user-configured)
- Use file-based communication
- Perform REAL validation before claiming completion
**You MUST NOT:**
- Write or edit ANY code files directly
- Skip quality gates
- Exceed iteration limits without user approval
- Pass large content through Task prompts
- Claim completion without screenshot evidence
- Accept "should work" or "tests pass" as proof
</orchestrator_role>
<real_validation_principle>
**Unit tests are NOT proof of working code.**
Real validation means:
- Deploy the application (dev server)
- Navigate browser to actual URL
- Take screenshot (compare to design if available)
- Perform real user actions (click, fill, submit)
- Verify expected behavior (redirects, content changes)
- Save evidence (screenshots, flow recordings)
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.
- yesterday First seen · 1,911 lines · 10 tokens per session scan A 092d2b4f2d46
feature is a command published in the GitHub repository MadAppGang/claude-code (279 stars, last pushed 5mo ago), licensed MIT. It adds 10 tokens to every session and 14,921 once invoked, about $0.0001 per session on Opus 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 commands, from other repositories
playwright
Write or update Playwright E2E tests following project conventions.
e2e
使用 Playwright 对 Web UI 进行端到端测试（支持视频录制、Trace 录制、控制台/网络日志捕获）.
paul:verify
Guide manual user acceptance testing of recently built features.
plan-spec
仅在人工明确安排 UI 自动化时，根据任务、产品规则或原型生成独立行为 Spec.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
ui-snapshot.template
This prompt was authored for Claude-style slash workflows. In Codex runtime, adapt tool calls as follows.