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 allemaar/open-skills --skill insight-adversarialgit clone --depth 1 https://github.com/allemaar/open-skillsWrote 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/allemaar/open-skills/insight-adversarial)<a href="https://agentmods.dev/skills/allemaar/open-skills/insight-adversarial"><img src="https://agentmods.dev/badge/skills/allemaar/open-skills/insight-adversarial/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/allemaar/open-skills/insight-adversarial"><img src="https://agentmods.dev/badge/skills/allemaar/open-skills/insight-adversarial.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00176 | $0.01639 |
| Opus 5 | $0.00088 | $0.00820 |
| Sonnet 5 | $0.00035 | $0.00328 |
| Haiku 4.5 | $0.00018 | $0.00164 |
Grade A, and why
insight-adversarial 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 12d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/insight-adversarial
Structured adversarial review that assembles a panel of distinct critic personas, runs each through a focused pass, and synthesizes a verdict. Personas are selected based on the nature of the target — a YON workflow gets different critics than a business plan or a UX design.
Structured execution spec:
protocol.yon. Read it for the canonical rules and step sequence; this file is explanation. The two must stay in sync — if you edit one, update the other and refresh the@STAMPdate.
Caller Options. Before executing, run the Caller Options protocol (
caller-options/SKILL.md): triage this invocation for material optionality across the venues and modes declared in front-matter; if one path clearly dominates, proceed silently; otherwise surface the options to the caller. A resolved-invocation marker means COP already ran — execute the fixed combination directly, do not re-enter COP.
How Passes Work
| Pass | What Happens |
|---|---|
| Pass 1 | Classify the target → select 4–5 personas → run each → synthesize verdict |
| Pass 2 | Check against Pass 1 findings → add 2–4 new personas not used before → re-run → delta verdict |
| Pass 3 | Final wrap-up. If all critical issues resolved or within params, declare done. Otherwise escalate. |
State tracking: at the top of each pass, list which personas were used in prior passes. New passes must not reuse them.
Step 1 — Target Classification
Before selecting personas, classify the target:
| Target Type | Key Question | Primary Risk |
|---|---|---|
plan |
Is this executable? | Scope drift, undeclared deps |
design / architecture |
Does structure match intent? | Coupling, ownership violations |
yon-workflow |
Does this execute correctly end-to-end? | Gate logic, step ordering, missing catches |
document / spec |
Is this complete and unambiguous? | Gaps, contradictions, assumed context |
idea / concept |
Is the core premise sound? | Wrong problem, premature solution |
business / product |
Is this viable and desirable? | Market assumptions, cost blindness |
code / implementation |
Is this correct and maintainable? | Edge cases, coupling, test coverage |
ui / ux |
Will real users succeed with this? | Flow breakdowns, mental model gaps |
What ships with it
2 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.
- 12d ago First seen · 136 lines · 176 tokens per session scan A 859be2f64f65
insight-adversarial is a skill published in the GitHub repository allemaar/open-skills (14 stars, last pushed 2d ago), licensed Apache-2.0. It adds 176 tokens to every session and 1,639 once invoked, about $0.0009 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
continual-learning
Nightly refinement of an existing per-repo review-style prompt using this reviewer's own finding outcomes. Read confirmed (resolved-by-commit / thumbs-up) and dismissed (thumbs-down) findings, promote the bug patterns the team actually fixes, demote the false-positive patterns, reconcile against the current prompt…
review-work
Post-implementation gate review: run manual QA on the real surface yourself, then launch ONE gate reviewer (never a panel) to audit goal, constraints, code quality, security, missed context, and QA evidence. Use before a PR handoff or when the user explicitly asks to review completed work.
security-review
Perform a focused security review of pending git changes to identify high-confidence security vulnerabilities with real exploitation potential. Use this skill when the user asks for a security review, security audit, vulnerability scan, or wants to check pending changes on a branch for security issues before merging.…
huggingface-llm-trainer
Train or fine-tune language models with TRL or Unsloth on Hugging Face Jobs, including SFT, DPO, GRPO, reward models, and GGUF conversion. Use for cloud LLM training; use huggingface-vision-trainer for vision tasks.
feature-dev
Guide a feature implementation through a structured seven-phase workflow with deep codebase understanding, clarifying questions, parallel architecture design, and quality review. Use this skill when the user asks to build a new feature, add functionality, or wants a methodical approach to implementation rather than…
make-pr-easy-to-review
Prepare PRs for review by cleaning noisy history, improving PR descriptions, and adding reviewer guidance without changing code behavior. Use for "make this easy to review", "tidy this PR", "clean up commits", or "annotate the diff".