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 daedalus/fuzzer --skill crash-analysisgit clone --depth 1 https://github.com/daedalus/fuzzerWrote 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/daedalus/fuzzer/crash-analysis)<a href="https://agentmods.dev/skills/daedalus/fuzzer/crash-analysis"><img src="https://agentmods.dev/badge/skills/daedalus/fuzzer/crash-analysis/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/daedalus/fuzzer/crash-analysis"><img src="https://agentmods.dev/badge/skills/daedalus/fuzzer/crash-analysis.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00162 | $0.02015 |
| Opus 5.5 | $0.00065 | $0.00806 |
| Sonnet 5.5 | $0.00032 | $0.00403 |
| Haiku 4.5 | $0.00016 | $0.00201 |
Grade C, and why
crash-analysis scanned grade C with 1 finding 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 15d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
> Mistake: running ``rm -rf crashes/*`` between analysis sessions. How it starts
The opening of the file, as written. The whole thing — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Crash Analysis
Reproduce, classify, and attribute crash artifacts from fuzzing campaigns using the project's sidecar format and existing analysis tooling.
When to Use This Skill
- "analyze this crash"
- "why did it segfault"
- "build a regression test for this crash"
- "generate a FINDINGS.md entry for this bug"
- "root cause this SIGFPE"
- "find the mutation site"
- "ASAN heap-buffer-overflow" / "UAF" / "UAF in libfoo"
- "crash forensics"
- "crash report" / "crash artifact"
- "minimize this crash input"
Crash Artifact Format
The fuzzer writes a quintet of sidecar files per crash (via save_crash in
adapters/filesystem.py):
crashes/
crash_<ts>_<cluster_id>_<san>_<err>.bin ← raw input bytes (only file counted as "crash")
crash_<ts>_<cluster_id>_<san>_<err>.txt ← human-readable report (target, returncode, GDB replay)
crash_<ts>_<cluster_id>_<san>_<err>.json ← same fields, machine-readable (CrashMetadata.to_dict())
crash_<ts>_<cluster_id>_<san>_<err>.hex ← xxd-style hexdump + text repr
crash_<ts>_<cluster_id>_<san>_<err>.sh ← base64 reproducer script
<san>_<err> is e.g. addr_heapbufferoverflow, or sig_signal11 without a
sanitizer report (_crash_name_parts).
The .bin is the authoritative input; the .txt and .json carry the
same triage fields (sanitizer, error_type, fault_addr, frames, registers,
gdb_replay, nearest_corpus, raw_stderr) — .txt for humans, .json for
dashboards and downstream tooling. The two are written together by
save_crash; treat them as a single artifact.
Core Methodology
1. Discover
Glob crashes/*.bin to enumerate inputs. Each .bin already has a
companion .json sidecar with the parsed triage fields — read those
directly:
jq '{returncode, sanitizer, error_type, fault_addr}' crashes/*.json
The .json is written by save_crash and mirrors CrashMetadata.to_dict()
(sanitizer, error_type, fault_addr, frames, registers, gdb_replay,
nearest_corpus_file, raw_stderr, target_sha256, returncode). No re-parse of the
.txt is needed.
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.
- 15d ago Changed · +3 lines 52e506378b5d
- 1mo ago First seen · 180 lines · 162 tokens per session scan C 6dd8e5fca8be
crash-analysis is a skill published in the GitHub repository daedalus/fuzzer (45 stars, last pushed today), licensed MIT. It adds 162 tokens to every session and 2,015 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
langsmith-observability
LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.
experimental-code-coverage-local-debugger
Runs code coverage locally via Universal Test Runner (UTR) or helper scripts, mimicking LUCI trybots. Activate when CQ tryjobs fail or underreport coverage, to test local GN/recipe repairs before uploading, or to debug hermetic crashes.
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
cli-e2e
Write, modify, or debug Docker-based Composio CLI end-to-end tests under ts/e2e-tests/cli, including binary invocation, fixture isolation, output assertions, and package manifests. Use for CLI E2E test suites only; use cli-command for CLI source implementation.
adversarial-reviewer
Stress-test a code change for concrete correctness defects, unsafe assumptions, and failure modes. Use when the user explicitly asks for an adversarial review, a deep bug hunt, or to tear a change apart, or to check your own change before submitting it. Do not use for ordinary review, style feedback, or general…