القائمة
Detect candidate fuzz targets and emit CI recipes for libFuzzer/AFL/cargo-fuzz/atheris/fast-check harnesses, plus OSS-Fuzz integration patterns
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/jmagly/aiwg --skill fuzzing-in-ciيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
WCAG accessibility analysis for color palettes including contrast ratios, compliance checking, and remediation suggestions. Use when user needs to verify colors meet accessibility standards.
Generate, analyze, compare, export, and suggest color palettes using color theory. Use when user asks about colors, palettes, color schemes, or needs help choosing colors for a project.
Research current color trends from Pantone, architecture, film, and design. Use when user asks about trending colors, popular palettes, or wants research-backed color inspiration.
استنادا إلى تصنيف SOC المهني
عرض SKILL.md
| namespace | aiwg |
| name | fuzzing-in-ci |
| platforms | ["all"] |
| description | Detect candidate fuzz targets and emit CI recipes for libFuzzer/AFL/cargo-fuzz/atheris/fast-check harnesses, plus OSS-Fuzz integration patterns |
| requires | [{"toolchain":"at least one supported language detected (C, C++, Rust, Python, Node)"},{"ci-platform":"known CI platform (Gitea Actions, GitHub Actions, GitLab CI)"}] |
| ensures | [{"harness-scaffolds":"starter fuzz harness(es) at .aiwg/security-engineering/fuzzing/{language}/"},{"ci-recipe":"PR-gating short-form fuzz job(s) at .aiwg/security-engineering/fuzzing/{ci-platform}/"},{"long-form-guide":"documentation for OSS-Fuzz integration and dedicated-runner patterns"}] |
| errors | [{"no-supported-language":"project has no language this skill can wire fuzzers for"},{"no-candidate-targets":"skill scanned for fuzz-worthy functions but found none; suggest manual seeding"}] |
| invariants | ["PR-gating fuzz runs are bounded (default 2 minutes per target) — never blocks PR indefinitely","long-form fuzzing is documented separately and never runs inline in PR jobs"] |
| script | {"entrypoint":"scripts/emit.mjs","runtime":"node","cwd":"project-root","argsHint":"[--language c|cpp|rust|go|python|node|auto] [--ci gitea|github|gitlab|auto] [--seconds-per-target N] [--coverage] [--oss-fuzz]"} |
| commandHint | {"argumentHint":"[--language c|cpp|rust|go|python|node|auto] [--ci gitea|github|gitlab|auto] [--seconds-per-target N] [--coverage] [--oss-fuzz]","allowedTools":"Read, Write, Bash, Glob, Grep","model":"haiku","category":"security","orchestration":false,"modelRole":"efficiency","modelTier":"economy"} |
You are the Fuzzing Integration Engineer — identify functions that take untrusted input, scaffold fuzz harnesses for them, and wire short-form PR-gating fuzz jobs that complement sanitizer-enabled builds.
"Coverage-guided fuzzing finds bugs the test suite never imagined." Fuzzers are the highest-leverage way to find input-handling bugs in parsers, deserializers, and protocol decoders. Wired into CI on every PR (with a small budget — minutes, not hours), they catch regressions immediately. Long-form fuzzing (OSS-Fuzz, dedicated runners) finds the deeper bugs over days and weeks.
| Language | Fuzzer | Property-based alternative |
|---|---|---|
| C / C++ | libFuzzer (Clang), AFL++ | — |
| Rust | cargo-fuzz (libFuzzer-backed), AFL (afl.rs) | proptest, quickcheck |
| Python | atheris (libFuzzer-backed) | Hypothesis |
| Node.js | jazzer.js (libFuzzer-backed) | fast-check |
| Java/JVM | jazzer (libFuzzer-backed) | jqwik |
Cycle-2 additions: Go (native go-fuzz), Swift, Ruby.
Same lib/toolchain-detect.sh helper as sanitizer-in-ci.
Heuristics for functions that benefit most from fuzzing:
bytes/&[u8]/Buffer inputparse_*, deserialize_*, decode_*, unmarshal_*from_str / from_bytes constructorsThe skill scans declarations using ripgrep:
# C/C++ — look for functions taking const uint8_t*/size_t pairs
rg -n --type c -- cpp -e src/
rg -n -- rust -e src/
rg -n -- py -e src/
rg -n -- js -- ts -e src/
Report candidates to the operator for confirmation before scaffolding harnesses.
Reference emitter:
agentic/code/frameworks/security-engineering/skills/fuzzing-in-ci/scripts/emit.sh \
--language auto --ci auto --seconds-per-target 120
.aiwg/security-engineering/fuzzing/c/fuzz_parse.c:
// LLVMFuzzerTestOneInput is the entry point libFuzzer expects.
// Compile with: clang -fsanitize=fuzzer,address,undefined -o fuzz_parse fuzz_parse.c parse.c
#include <stdint.h>
#include <stddef.h>
#include "parse.h"
int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) {
// Constrain input shape if your parser has a minimum
if (size < 4) return 0;
parse_result_t *r = parse_buffer(data, size);
if (r) {
parse_result_free(r);
}
return 0;
}
.aiwg/security-engineering/fuzzing/c/seed-corpus/ — populate with a few representative valid inputs.
fuzz/fuzz_targets/fuzz_parse.rs (cargo-fuzz layout):
#![no_main]
use libfuzzer_sys::fuzz_target;
use mycrate::parse;
fuzz_target!(|data: &[u8]| {
let _ = parse(data); // ignore Result; we want it to not panic/UB
});
Project-side: cargo install cargo-fuzz then cargo fuzz init && cargo fuzz add fuzz_parse.
.aiwg/security-engineering/fuzzing/python/fuzz_parse.py:
import atheris
import sys
with atheris.instrument_imports():
from mypkg import parse
def TestOneInput(data: bytes) -> None:
try:
parse(data)
except (ValueError, mypkg.ParseError):
pass # expected on invalid input
atheris.Setup(sys.argv, TestOneInput)
atheris.Fuzz()
.aiwg/security-engineering/fuzzing/node/fuzz_parse.js:
const { parse } = require("../../../src");
module.exports.fuzz = function (data /* Buffer */) {
try {
parse(data);
} catch (e) {
// Only catch the parser's documented error class; let everything else propagate
if (!(e instanceof require("../../../src").ParseError)) throw e;
}
};
Short-form fuzzing on every PR. Default budget: 2 minutes per target, with a corpus that grows over time.
# .aiwg/security-engineering/fuzzing/github/c.yaml
name: Fuzz (PR)
on:
pull_request:
push:
branches: [main]
jobs:
libfuzzer:
runs-on: ubuntu-latest
timeout-minutes: 10
steps:
- uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
- name: Install clang
run: apt-get update && apt-get install -y clang
- name: Cache fuzz corpus
uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3
with:
path: .aiwg/security-engineering/fuzzing/c/corpus
key: fuzz-c-corpus-${{ github.run_id }}
restore-keys: fuzz-c-corpus-
- name: Build fuzz target
run: clang -fsanitize=fuzzer,address,undefined -O1 -g -o fuzz_parse \
.aiwg/security-engineering/fuzzing/c/fuzz_parse.c src/parse.c
- name: Fuzz for 120 seconds
run: |
mkdir -p .aiwg/security-engineering/fuzzing/c/corpus
./fuzz_parse -max_total_time=120 \
.aiwg/security-engineering/fuzzing/c/corpus \
.aiwg/security-engineering/fuzzing/c/seed-corpus
- name: Upload crashes
if: failure()
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882 # v4.4.3
with:
name: fuzz-crashes-c
path: crash-*
.aiwg/security-engineering/fuzzing/OSS-FUZZ.md:
# OSS-Fuzz Integration
OSS-Fuzz runs continuous fuzzing on Google infrastructure for free, for qualifying open-source projects.
## Eligibility
- Open-source project with significant impact (usage in production, dependency of widely-used software, etc.)
- Submit via https://github.com/google/oss-fuzz/blob/master/docs/getting-started/new_project_guide.md
## Required files
For a project named `myproject`, OSS-Fuzz needs:
1. `projects/myproject/project.yaml` — homepage, language, sanitizers
2. `projects/myproject/Dockerfile` — build environment
3. `projects/myproject/build.sh` — compiles fuzz targets
Example:
```yaml
# project.yaml
homepage: "https://example.com/myproject"
language: c
primary_contact: "security@example.com"
auto_ccs:
- "another-maintainer@example.com"
sanitizers:
- address
- undefined
- memory # optional, requires MSan-clean deps
# Mirror what OSS-Fuzz does locally:
python infra/helper.py build_image myproject
python infra/helper.py build_fuzzers --sanitizer address myproject
python infra/helper.py run_fuzzer myproject fuzz_parse
### Phase 6: Property-based testing recipes
For languages or codebases where coverage-guided fuzzing is awkward (high-level Python, async-heavy JS, JVM with lots of reflection), property-based testing is often easier to adopt and complementary.
`.aiwg/security-engineering/fuzzing/python/property_test.py`:
```python
from hypothesis import given, strategies as st
from mypkg import parse, serialize
@given(st.binary(min_size=0, max_size=10_000))
def test_parse_never_panics(data):
try:
parse(data)
except mypkg.ParseError:
pass
@given(st.text())
def test_roundtrip(s):
assert parse(serialize(s)) == s
.aiwg/security-engineering/fuzzing/node/property_test.js (fast-check):
const fc = require("fast-check");
const { parse, serialize } = require("../../../src");
test("roundtrip", () => {
fc.assert(fc.property(fc.string(), (s) => parse(serialize(s)) === s));
});
test("parse handles arbitrary bytes", () => {
fc.assert(
fc.property(fc.uint8Array(), (bytes) => {
try { parse(Buffer.from(bytes)); } catch (e) { /* expected */ }
})
);
});
sanitizer-in-ci — fuzzers find inputs that trigger UB; sanitizers detect the UB. Pair them.ci-action-pinning — all action SHAs are pinned.scripts/emit.sh emits starter harnesses under .aiwg/security-engineering/fuzzing/{language}/..aiwg/security-engineering/fuzzing/{ci-platform}/.OSS-FUZZ.md and merge_corpus.sh placeholders..aiwg/security/curl-checklist-gap-analysis.md row 16