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/happier-dev/happier/attack-conclusionnpx skills add happier-dev/happier --skill attack-conclusiongit clone --depth 1 https://github.com/happier-dev/happierWhat 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 | $0.01588 |
| Opus 5 | $0.00044 | $0.00794 |
| Sonnet 5 | $0.00018 | $0.00318 |
| Haiku 4.5 | $0.00009 | $0.00159 |
Grade A, and why
attack-conclusion 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Attack Your Conclusion
Before handing over a conclusion, switch roles completely: you are no longer the author defending it, you are the reviewer paid to break it, with the same energy spent building it. Full doctrine: docs/agent-craft.md §6 and §8.
The test of whether you actually switched roles: did you go looking for evidence that would change your mind, or only re-inspect the evidence that formed the conclusion?
The standard attacks — in order of cheapness, each as a runnable check
- Alternative cause or falsifier. Ask what else could explain the same evidence. When the evidence supports a materially different candidate, name it and run the cheapest discriminating observation; when the mechanism is directly established, do not manufacture a second hypothesis—identify and run the cheapest observation that could falsify the conclusion instead.
- Neighboring cases. The fix works for the reproduced case. Run the case next door: the empty list, the second invocation, the other platform, the resumed session, the concurrent caller.
- Blast radius. What consumes what you changed? Search callers, readers, subscribers, tests, serialized forms. "Nothing else uses this" is a claim — re-derive it, don't assert it.
- Environment gap. Does the conclusion survive where the code actually runs, or only in the harness? Host tests encode the same assumptions the author had. For user-visible behavior, use the risk-appropriate browser/device gate in
skills/happier-testing. - Hypothesis lock. Are you explaining the evidence, or explaining your first hypothesis? Re-read the raw evidence pretending you just arrived and have no favorite.
Run the cheap attacks; an attack that is just worry is not an attack. If you cannot state what would falsify the conclusion, it is not a conclusion yet — it is a preference.
Architecture-impact attack
Apply root Scope-preserving solution economy during this attack: preserve the complete feature outcome, challenge unsupported implementation machinery rather than the feature itself, and try folding behavior into the canonical owner before accepting a split-brain or parallel path.
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 · 67 lines · 89 tokens per session scan A d18e24e86c9d
attack-conclusion is a skill published in the GitHub repository happier-dev/happier (1,608 stars, last pushed today), licensed MIT. It adds 89 tokens to every session and 1,588 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 2026-08-30.
Other skills, from other repositories
maple-preview
Use when a MapleStory asset request resolves to more than one plausible candidate: a look described by colour, style, or mood instead of an exact name or ID (a black cap, a white tee, a monster that suits this map), a maple-lookup search returning several close matches, or the user asking to preview, compare, or…
balance-tuning
Reviews and tunes game balance across difficulty and progression curves, session pacing, option or strategy trade-offs, and reward economies. Use when players report that a game is too hard, easy, short, long, or grindy; a weapon, upgrade, or strategy dominates; damage, health, spawn, drop, price, or progression…
cross-device-check
Audits a game's UI and interaction across supported viewports, aspect ratios, pixel densities, orientations, and input capabilities. Use when mobile, tablet, desktop, controller, keyboard, pointer, or touch support is being checked; UI clips, overlaps, blurs, or overflows; controls are unreachable or hard to use…
determinism-audit
Audits nondeterministic game behavior and flaky execution. Use when the same initial state and inputs produce different outcomes, a test fails intermittently or only under load, behavior differs by frame rate or machine, a replay diverges, or lockstep peers desynchronize. Establishes the required determinism boundary…
game-feel
Use when a game's controls feel stiff, floaty, heavy, slippery, unresponsive, or unfair; when players miss jumps they thought they made, snag on platform corners, lose inputs pressed during another action, or get hit by grazes; when a platformer, runner, or action game is being built or polished and its input and…
game-tutorial
Audits, adds, or reworks a game's first-time-player tutorial. Use when players cannot identify the controls, goal, failure conditions, or a non-obvious rule; new players quit before completing the core loop; onboarding presents too much at once; or an explicit tutorial or first-run improvement is requested. Separates…