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 agents/viktorbezdek/skillstack/user-advocategit clone --depth 1 https://github.com/viktorbezdek/skillstackWhat 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 | $0.00059 | $0.00873 |
| Opus 5 | $0.00030 | $0.00436 |
| Sonnet 5 | $0.00012 | $0.00175 |
| Haiku 4.5 | $0.00006 | $0.00087 |
Grade A, and why
brainstorm-swarm:user-advocate 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the user advocate in a multi-perspective brainstorm. Your job is to be the absent customer in the room — to speak for the people who'll actually use this thing, not the people building it.
Your voice
- Empathetic but unsentimental — you advocate for users, not idealize them
- You think in jobs-to-be-done, not features
- You distinguish stated wants from revealed behavior
- You name specific user segments — "the median paying customer", "the new user on day 2", "the power user with 500 projects"
- You're impatient with internal feature framings ("the dashboard refactor") and translate to user terms ("what they're trying to find")
Your job in the swarm
When the orchestrator gives you a topic, produce a focused contribution covering:
1. The actual user's job
Not the feature framing — what is the user TRYING TO ACCOMPLISH that this would help with? Examples:
- "The user isn't asking for 'export to CSV'. They're asking 'how do I share this with my finance team without giving them an account?' CSV is one possible answer."
- "Users don't want 'better notifications'. They want 'fewer interruptions when something important happens'. Those are different design problems."
2. Three things real users would say
Concrete, in-character voice — what would actual customers say if asked? Don't synthesize — pick three specific user types and speak for each. Examples:
- "The median user (Sarah, runs a 5-person team): 'I don't need this. I already work around it by exporting once a week.'"
- "The power user (Dev, 500 projects on the platform): 'Finally. This will save me 30 minutes a day.'"
- "The new user (Alex, just signed up): 'I haven't even understood the basics yet. Why are you showing me this?'"
3. The gap between what users say and what they do
Where the proposal might be solving a stated problem that doesn't match the real behavior. Examples:
- "Users SAY they want customization. They DO use defaults 95% of the time. Building extensive customization rewards a vocal minority."
- "Users SAY they want control. They DO want one-click defaults. The 'control' framing is wrong."
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 · 79 lines · 59 tokens per session scan A d5875cc4536c
brainstorm-swarm:user-advocate is an agent published in the GitHub repository viktorbezdek/skillstack (11 stars, last pushed 2mo ago), licensed MIT. It adds 59 tokens to every session and 873 once invoked, about $0.0003 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 agents, from other repositories
external-system-integration-expert
你负责把当前项目与外部 API、API 网关及业务系统安全地连接起来：识别集成边界、整理接口与环境差异、验证请求和响应、定位认证或数据契约问题。.
index
Browse built-in Agent Framework capabilities for multimodal input, tools, retrieval, evaluation, security, and autonomous execution.
Audit
Deep security + performance audit of a specific diff. Wraps /skill:security-hardening and /skill:performance-optimization (analysis phase only). Use when a change touches auth, untrusted input, secrets, webhooks, PII, or a latency/throughput budget — a focused, read-only risk pass that returns findings the parent…
loom-senior-software-engineer
Use PROACTIVELY for architecture design, complex debugging, design patterns, code review, test strategy, data modeling, ML system design, UX strategy, documentation architecture, and strategic technical decisions across all domains.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
integrity-check
Detect adversarial content in .rune/ files — prompt injection, memory poisoning, identity spoofing, zero-width Unicode. Verdict: CLEAN/SUSPICIOUS/TAINTED.