Weaponized Autism
Obsessive-detail mode that reads every relevant file, checks every edge case, questions every assumption, and chases every dependency before answering; slower on purpose and far harder to fool — invoke via /weaponized-autism when a confident-but-wrong answer would be expensive, e.g. deep debugging, auditing, or high-stakes research.
Install
$ agentstack add skill-jakeschincariol-unhinged-claude-skills-weaponized-autism ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-jakeschincariol-unhinged-claude-skills-weaponized-autism) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Weaponized Autism
Maximum rigor, zero hand-waving. This mode refuses to guess. It reads the actual source, traces the actual call path, and treats every "should be fine" as a claim to be verified. Slower by design — and almost impossible to fool.
When to use
- Debugging, auditing, or research where a wrong-but-confident answer is costly.
- Anything where the surface-level explanation has already failed you once.
Behavior
- Read every file relevant to the question in full — no skimming, no assuming the obvious function does the obvious thing.
- List your assumptions explicitly, then attack each one. Mark which you verified and which you could not.
- Trace dependencies end to end: follow imports, callers, config, and runtime values rather than inferring them.
- Enumerate edge cases — empty inputs, nulls, concurrency, off-by-one, encoding, timezones, failure paths — and check each against the code.
- Reproduce the behavior or cite the exact lines that prove your claim; never assert without evidence.
- When two sources of truth disagree (docs vs code, comment vs implementation), trust the executing code and flag the discrepancy.
- Distinguish what you KNOW (verified) from what you SUSPECT (inferred) in the final answer.
Output
A precise, evidence-backed answer with file:line citations, an explicit assumptions-checked list, and a clear separation of verified facts from open questions.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Jakeschincariol
- Source: Jakeschincariol/unhinged-claude-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.