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/outlinedriven/odin-claude-plugin/offensenpx skills add OutlineDriven/odin-claude-plugin --skill offensegit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote 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/outlinedriven/odin-claude-plugin/offense)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/offense"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/offense.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 | $0.00089 | $0.01310 |
| Opus 5 | $0.00044 | $0.00655 |
| Sonnet 5 | $0.00018 | $0.00262 |
| Haiku 4.5 | $0.00009 | $0.00131 |
Grade A, and why
offense 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 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.
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 — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Offense
Radical codebase overhaul, invoked by a human only. Demolish-and-re-derive: rewrite subsystems from their contracts instead of patching, delete rather than deprecate, collapse special cases into the general case, and migrate every caller in the same change. The stance is offensive — tear the broken structure out and re-derive it clean — not defensive patching that leaves the old shape half alive.
Contract
| Field | Bound contract |
|---|---|
| Trigger | A human invokes an overhaul: "overhaul", "rebuild this subsystem", "rewrite it from scratch", or equivalent. Never self-initiated. |
| Authority | Modify only the invoked scope, an exact disclosed set of repository paths governed by version control. Create atomic commits and use version control to recover from an invalid increment. Stop and ask before removing any observable surface consumers depend on. |
| Side effect | Atomic compilable commits that demolish the old structure and re-derive it from its contracts; the general case absorbs every special case; every caller migrates in the same change; no shim, alias, deprecated path, or feature flag survives the cutover. |
| Done | The invoked scope is rebuilt from its contracts, every caller is migrated, no special-case branch or deprecated path remains, the feature works end to end, and the applicable verifier is green. |
Inputs
Supply the invoked scope (the subsystem, module set, or surface to overhaul), the contracts that govern it, and the permitted path set. Supply required verification commands when the repository does not define them. An implementation plan or known invariants are optional; derive them from the existing contracts when absent, but do not invent requirements.
Procedure
- Bound the invoked scope. List the exact repository paths in the overhaul and the observable surfaces it exposes (public APIs, CLI flags, config keys, file formats, wire shapes). Stop and ask before removing any observable surface consumers depend on; never silently narrow scope. Done when: the exact path set and observable surfaces are listed, and removal of any consumer-facing surface is explicitly approved.
- Re-derive the contract. Read the existing implementation and extract the contract the subsystem should have held — its invariants, trust boundaries, and the general case that absorbs every special case. Discard the old structure's accidents; keep only the contract. Done when: the contract is written with invariants, trust boundaries, and the general case named.
- Partition into atomic increments. Each increment must preserve the named invariants, compile, and leave applicable existing behavior verifiable. Order increments so the contract is re-derived before callers migrate onto it. Done when: the work is partitioned into increments that each compile and preserve invariants.
- Demolish and re-derive. Implement the general case directly from the contract. Encode invariants in types or construction where practical; otherwise validate at the boundary and return or raise the repository's explicit error form. Delete the old structure rather than deprecating it; collapse every special-case branch into the general rule; do not leave a shim, alias, deprecated path, or feature flag for the old shape. Done when: the general case is implemented, the old structure is deleted, and no special-case branch, shim, alias, or dark flag remains.
- Migrate every caller in the same change. Update every callsite to the new contract in the same change that introduces it; never leave a caller on a removed surface. If a caller cannot migrate within the disclosed path set, stop and ask — do not widen scope implicitly and do not leave a half-migrated state. Done when: every caller is migrated and no reference to a removed surface remains.
- Verify each increment. After each increment, run the narrowest applicable compile and behavioral checks that exercise its changed contract. If they pass, commit that one logical change with a descriptive message. Revalidate all named invariants affected by the next increment before continuing. Done when: each increment is checked, committed, and its invariants revalidated for the next increment.
- Exercise end to end and run the final verifier. Confirm the changed-path set is exactly the disclosed set, that no special-case branch, shim, alias, deprecated path, or uncommitted partial increment remains, and that the feature works end to end. Done when: end-to-end behavior is exercised, the final verifier is green, and the changed-path set matches the disclosed set with no remnants.
What ships with it
1 file 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.
- yesterday First seen · 40 lines · 89 tokens per session scan A 2d3e8acd13cc
offense is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed yesterday), licensed Apache-2.0. It adds 89 tokens to every session and 1,310 once invoked, about $0.0004 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 2026-09-04.
Other skills, from other repositories
java-clean-arch
Reviews or implements Clean Architecture / Hexagonal Architecture (Ports & Adapters) and DDD tactical patterns for Java projects. Use when user asks to "apply clean architecture", "implement hexagonal architecture", "add ports and adapters", "apply DDD", "refactor to clean arch", "review architecture", or "add value…
java-adr
Creates, lists, and manages Architecture Decision Records for Java projects. Use when user asks to "create an ADR", "document this decision", "write an architecture decision", "add ADR", "list decisions", "show ADRs", or "record this architectural choice".
java-test
Generates JUnit 5 and Mockito unit tests or Testcontainers integration tests, auto-detecting project setup. Use when user asks to "write tests", "generate tests", "add unit tests", "create test class", "test this service", or "write integration tests".
java-jpa
Reviews Spring Data JPA for N+1 queries, fetch strategies, projections, and Specifications. Use when user asks to "review JPA", "check for N+1", "JPA performance", "review my entities", "check fetch strategy", or "review my repositories".
java-logging
Reviews Java logging for SLF4J best practices, MDC context, structured logging, and PII safety. Use when user asks to "review logging", "check my logs", "logging review", "is my logging correct", "MDC setup", or "check for PII in logs".
java-security
Reviews or implements Spring Security configuration — JWT authentication, OAuth2, method-level security, CORS, and CSRF. Use when user asks to "add authentication", "secure this API", "implement JWT", "configure Spring Security", "add OAuth2 login", "protect endpoints", or "review security config".