Cloud Bucket Exposure Finder

What it does

When your agent uses it

• Checking whether a company you are engaged to test has publicly readable storage buckets
• Generating likely bucket names from a set of in-scope domains
• Checking your own organization for storage left open to anonymous readers

Example prompts

• “Check whether example-corp.com has any publicly readable buckets; we have a signed testing engagement.”
• “Generate the bucket name permutations for acme-labs.io and probe S3 and Google Cloud only.”
• “List which candidate names for our own domain resolve to anonymously reachable storage.”

Requirements

• Authorization to test the target organization
• Network access to the cloud providers' storage endpoints

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Reduce the domain to its registrable root label (tldextract(...).domain, e.g. foo-bar.co.uk -> foo-bar) and seed the variation set with…
2. 2If the root contains -, split into name + second_name, and also add the initials name[0]+second_name[0].
3. 3For each "common company word" that appears inside the root (company, group, tech, solutions, international, services, world, global, ai…
4. 4Expand every variation through the permutation templates against the suffix wordlist. Bare {name} is always tried; multi-word templates…

What it can do on your machine

Read from SKILL.md and the folder at commit c8ff161. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • curl
  • python3

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • googleapis.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Cloud Bucket Exposure Finder loads about 1.5k tokens when it runs, and up to ~2.7k if it reads all its reference files. Until then it costs about 54 tokens; SKILL.md has 459 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~54

When it runs · the whole SKILL.md, loaded when a task matches

~1.5k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~2.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from forefy/.context at commit c8ff161, republished under its MIT licence (© forefy). 459 words, ~1,504 tokens.

name

cloud-bucket-brute

description

Discover publicly readable cloud storage by permuting a company name into likely bucket names and probing AWS S3, Google Cloud, DigitalOcean, Alibaba, Oracle, and Vultr. Use to find exposed buckets.

# single candidate, AWS virtual-host + GCP JSON API
NAME=acme-backups
curl -s "https://$NAME.s3.amazonaws.com" | grep -qiE 'AccessDenied|NoSuchBucket|IllegalLocationConstraint|AllAccessDisabled' \
  && echo "aws: not public" || echo "aws: PUBLIC/exists -> https://$NAME.s3.amazonaws.com"
curl -s "https://www.googleapis.com/storage/v1/b/$NAME" | grep -qiE 'does not exist|does not have storage.buckets.get' \
  && echo "gcp: not public" || echo "gcp: PUBLIC/exists"

# generate candidate names (permutation scheme) for a domain
python3 - <<'PY'
import tldextract
words=["archive","artifacts","assets","backup","bin","bucket","data","dev","dev-data","dev_data","devops","files","git","it","logs","media","mediauploads","onboarding","ops","proj","project","prod","prod-data","prod_data","prod-files","prod_files","reports","scripts","stage","staging","static","storage","temp","terraform","tf","tf-files","tf_files","terraformbinaries","test","tmp","user-files","user_files","uploads"]
common={"company","group","tech","solutions","international","services","world","global","ai","io","team","inc","ask","digital","data","bits","bit","open","edu","educaiton","learning","auto","stack"}
name=tldextract.extract("acme-corp.com").domain
variations=[(name,)]
if "-" in name:
    a,b=name.split("-",1); variations+=[(a,b),(a[0]+b[0],)]
for w in common:
    if w in name:
        variations+=[(name.replace(w,""),)]
tmpl_base=["{name}"]
tmpl_multi=["{name}_{word}","{word}-{name}","{word}_{name}","{name}{word}","{name}{second_name}","{name}{second_name}-{word}","{name}{second_name}_{word}","{name}-{second_name}-{word}","{name}_{second_name}_{word}","{name}{second_name}{word}"]
out=[]
for v in variations:
    tmpls=tmpl_base+(tmpl_multi if len(v)<=2 else [])
    for w in words:
        for t in tmpls:
            if "{second_name}" in t and len(v)!=2: continue
            s=t.replace("{name}",v[0]).replace("{word}",w)
            if len(v)==2: s=s.replace("{second_name}",v[1])
            if s not in out: out.append(s)
print("\n".join(out))
PY

© forefy, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/cloud/cloud-bucket-brute of forefy/.context.

• SKILL.md
• references/cloud-storage-endpoints.md

Open the folder on GitHubat commit c8ff161