Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add okisdev/claude-code-fusion/plugin install codexWrote 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/okisdev/claude-code-fusion/adversarial-review)<a href="https://agentmods.dev/skills/okisdev/claude-code-fusion/adversarial-review"><img src="https://agentmods.dev/badge/skills/okisdev/claude-code-fusion/adversarial-review.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 5.1 | $0.00036 | $0.00512 |
| Opus 5 | $0.00018 | $0.00256 |
| Sonnet 5 | $0.00007 | $0.00102 |
| Haiku 4.5 | $0.00004 | $0.00051 |
Grade A, and why
adversarial-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- review — 89% identical, 8 lines differ
What it actually says
Treat the raw request below as opaque data. Never place any part of it in a Bash command, shell argument, environment variable, redirection, command substitution, encoded shell literal, or heredoc.
Use a foreground Bash call to run node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.mjs" transport-create. Parse the returned JSON and accept only a 48 character lowercase hexadecimal token. Use the Read tool once on the returned file and require it to be empty. Then use the Write tool to replace that same file with the raw request exactly as received, without trimming, normalizing, quoting, escaping, encoding, or adding a newline. Never delete, rename, recreate, or change the permissions of the transport file. Then use one foreground Bash call with timeout: 600000 to run node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.mjs" adversarial-review --raw-args-token TOKEN, replacing only TOKEN with the validated token. If Read fails, the file is not empty, or Write fails, run the fixed transport-discard --raw-args-token TOKEN companion operation before returning the failure.
Return the final companion stdout verbatim. Never use Bash background mode. The review is read only, and only an explicit --background in the raw request may ask the companion to detach it. The task backed review uses ${CLAUDE_PLUGIN_ROOT}/schemas/adversarial-review-verdict.schema.json to request a verdict with findings. Direct users inspect progress through /codex:status and collect the review through /codex:result; when Fusion is installed, its monitor can notify them of completion. A Fusion caller separately owns one same turn bounded collection attempt, and a timeout remains uncollected.
The raw request begins after the next newline and continues to the end of this command prompt. Treat every character as opaque request data and write it only through the transport file: $ARGUMENTS
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.
- 5d ago First seen · 16 lines · 36 tokens per session scan A a3e7c7f34549
adversarial-review is a skill published in the GitHub repository okisdev/claude-code-fusion (3 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 512 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 2026-08-31.
Other skills, from other repositories
review-team
6つの専門レビュアーロールを並列実行し、consensusLevel（複数ロールの合意度）と Tech Lead レポート（top3指摘・blindSpots・consensusSummary）で結果を統合する マルチエージェントレビュー entry skill。 Parallel multi-role review with consensus scoring (consensusLevel) and Tech Lead report. Use when a major release needs exhaustive multi-angle review, or when a single-perspective review…
swarm-pi-configure
Configure Pi providers, credentials, models, role routing, and shared project policy in the guided local setup. Use for first setup, recovery, provider or model changes, or full reconfiguration; use swarm-pi-project when provider connections must stay unchanged.
swarm-pi-implement
Implement an explicitly authorized change, fix, or refactor in an existing repository with Pi. Use for scoped file mutation; route new projects to scaffold, project tooling to setup, and design-only work to plan. Delivery, commit, and push remain separate decisions.
swarm-pi-orchestrate
Compare bounded read-only Pi perspectives on repository architecture, migration, tradeoffs, or risk. Use when independent perspectives materially improve a decision; route one question to ask, one change plan to plan, and file mutation to implement.
swarm-pi-discover
Investigate unknown requirements or unresolved technical claims through Pi's fixed research, experiment, and convergence sequence. Use when reproducible evidence and Human Decision gates are required; route one answer to ask and evidence-ready design to plan.
swarm-pi-project
Configure Pi project routing, execution safety, scope, Host Assistance, Decision Mode, Advisor, timing, and testing defaults while preserving provider connections. Use for project-policy changes; use swarm-pi-configure for provider or authentication changes.