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 osama-raddad/FireCrasher --skill customize-recoverygit clone --depth 1 https://github.com/osama-raddad/FireCrasherWrote 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/osama-raddad/firecrasher/customize-recovery)<a href="https://agentmods.dev/skills/osama-raddad/firecrasher/customize-recovery"><img src="https://agentmods.dev/badge/skills/osama-raddad/firecrasher/customize-recovery/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/osama-raddad/firecrasher/customize-recovery"><img src="https://agentmods.dev/badge/skills/osama-raddad/firecrasher/customize-recovery.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 | $0.00053 | $0.00548 |
| Opus 5.5 | $0.00021 | $0.00219 |
| Sonnet 5.5 | $0.00011 | $0.00110 |
| Haiku 4.5 | $0.00005 | $0.00055 |
Grade A, and why
customize-recovery 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 2mo 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Customize the recovery experience
FireCrasher escalates recovery through three RecoveryLevels:
- RESTART_ACTIVITY — restart the crashing activity (occasional crash).
- GO_BACK — the activity keeps crashing, so go back to the previous screen.
- RELAUNCH_APP — nothing to go back to, so restart the whole app.
You can read the level before recovering and show appropriate UX, or force a level.
React to the recovery level
Inside onCrash, the evaluated level and the current activity are already
in scope. Call recover { … }; its lambda runs after recovery is kicked off.
installFireCrasher {
onCrash {
report(throwable) // see report-crashes skill
val context = activity ?: return@onCrash
when (level) {
RecoveryLevel.RESTART_ACTIVITY ->
// Quiet restart of the current screen — show a brief spinner.
recover {
Toast.makeText(context, "Recovering…", Toast.LENGTH_SHORT).show()
}
RecoveryLevel.GO_BACK ->
recover {
Toast.makeText(context, "Returning to the previous screen", Toast.LENGTH_SHORT).show()
}
RecoveryLevel.RELAUNCH_APP ->
recover {
Toast.makeText(context, "Restarting the app", Toast.LENGTH_LONG).show()
}
}
}
}
retryCount is also in scope if you want to distinguish the first attempt
from later ones (e.g. only show a dialog once):
if (retryCount <= 1 && level == RecoveryLevel.RESTART_ACTIVITY) {
// show a loading dialog
}
Force a specific level
If you know how you want to recover, pass the level explicitly instead of using the evaluated one:
onCrash {
recover(RecoveryLevel.RELAUNCH_APP) // always restart the app on any crash
}
Guidance
- Keep recovery UX fast and non-blocking — the user just hit a crash; a long modal makes it worse. A spinner or short toast is usually enough.
- Do the crash reporting before showing UX (see
report-crashes). - The recovery lambda runs on the main thread; don't do heavy work in it.
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.
- 2mo ago First seen · 73 lines · 53 tokens per session scan A a829f58f3fa6
customize-recovery is a skill published in the GitHub repository osama-raddad/FireCrasher (147 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 53 tokens to every session and 548 once invoked, about $0.0002 per session on Opus 5.5. 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.
customize-recovery compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| customize-recoveryosama-raddad/FireCrasher | Skill | 147 | 4mo ago | Apache-2.0 | A | 53 | Claude Code, Codex |
| mobile-designmanutej/luxor-claude-marketplace | Skill | 63 | 4mo ago | none | A | 23 | Claude Code |
| solopi-aialipay/SoloPi | Skill | 6.4k | yesterday | Apache-2.0 | A | 82 | Codex |
| argent-tv-interactsoftware-mansion/argent | Skill | 3.2k | yesterday | Apache-2.0 | A | 107 | Claude Code, Codex |
| analyzing-android-malware-with-apktoolmukul975/Anthropic-Cybersecurity-Skills | Skill | 36k | 2mo ago | Apache-2.0 | A | 84 | Claude Code |
| compose-multiplatformericrisco/rsc-harness | Skill | 503 | 2d ago | MIT | A | 80 | Claude Code, Codex |
| simvynpranshuchittora/simvyn | Skill | 442 | 9d ago | MIT | A | 63 | Codex |