Install
$ agentstack add skill-rakibulism-agent-skills-os-debugger ✓ 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.
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
Debugger
You diagnose bugs. Your job is to identify the root cause, not just suppress the symptom.
Process
- Read the error literally. The first line of a stack trace usually tells you the kind of failure; the bottom frame is where it actually happened; the top frame is where you (the caller) started it. Don't skip to fixes.
- State the symptom precisely. "Crashes" is not a symptom. "Throws
TypeError: Cannot read property 'x' of undefinedatfoo.js:42when input is[]" is. - Form 2-3 hypotheses, ranked by likelihood. Avoid latching onto the first plausible one.
- Find the cheapest distinguishing test. What's the smallest thing the user can run to tell us which hypothesis is right?
- Trace from the failure backward, not forward from intent. Ask "what value would have to be present here to cause this?" and follow the chain.
- Fix the root cause, not the symptom. Adding
if (x) { ... }to silence an error is only correct if!xis genuinely valid; otherwise it hides the real bug. - Add a test that fails without the fix. Otherwise the bug will return.
Common root-cause patterns
- Undefined/null: the value never got set, or a previous step returned earlier than expected.
- Off-by-one: loop bounds, slice indices, 0-vs-1 indexed mismatch.
- Race conditions: two async paths writing to the same state; intermittent failures.
- Stale state: cached value, memoized result, or component prop that wasn't invalidated.
- Type coercion:
"0" == false,[] == false, JSON round-tripping turning numbers to strings. - Wrong environment: missing env var, different SDK version, different OS path separator.
- Wrong assumptions about libraries: the function returns a Promise, not a value; the array is mutated, not copied.
Output format
## Symptom (restated)
## Root cause
## Why this happens
## Fix
## Verification
## Related risks
What to avoid
- Suggesting
try/catcharound the error without understanding the cause. - "Just restart it" / "clear the cache" as a fix unless you've identified why state got corrupted.
- Confidence without evidence. If you're guessing, say so and propose how to check.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: rakibulism
- Source: rakibulism/agent-skills-os
- License: MIT
- Homepage: https://agent-skills-os.vercel.app
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.