// MCP TOKEN COST

The Osint Terminal MCP server costs 22,880 tokens before the first call.

Every request your agent makes carries every tool definition this server exposes — context your code, documents and conversation can't use, mostly for tools the agent never calls. You don't need them all in the window, and you don't have to pay for them.

QUICK ANSWER The Osint Terminal MCP server's 474 tool definitions consume 22,880 tokens — 11% of a 200k context window, and 12× the median MCP server (1,862 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS tiktoken o200k_base · rank #505 of 12,722 measured servers · refreshed every build Method →

// CONTEXT WINDOW SHARE

What that costs before your agent starts working.

Tool definitions are overhead: they occupy context on every request and compete with your code, documents and conversation history for the same window.

200K WINDOW 11%

1M WINDOW 2.3%

Corpus context: Osint Terminal ranks #505 of 12,722 measured MCP servers by definition cost. The median is 1,862 tokens, p90 is 13,674, and the heaviest (Ainumbers Mcp Apps) is 442,047 — 221% of a 200k window on its own. New to this? See MCP token cost and context window in the glossary.

Token cost is one axis. See the risk picture across your whole stack →

// PER-TOOL BREAKDOWN

Where the 22,880 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 48 tokens per tool.

Tool	Category	Tokens	% of server
social_deep	Write	97	0.4%
correlate_sweep	Execute	83	0.4%
workup	Read	67	0.3%
intelx_email	Read	63	0.3%
person_sweep	Read	62	0.3%
leaklookup	Read	61	0.3%
base58	Read	60	0.3%
snusbase_hash	Read	60	0.3%
airquality	Read	59	0.3%
fbsearch	Read	59	0.3%
jarm	Read	59	0.3%
phone_deep	Read	59	0.3%
punycode	Read	59	0.3%
hibp	Read	58	0.3%
ip_deep	Read	58	0.3%
hibp_email	Read	57	0.2%
vuln_feed	Read	57	0.2%
breach_aggregator	Read	56	0.2%
disasters	Read	56	0.2%
doh	Read	56	0.2%
emerging_threats	Read	56	0.2%
imagechunks	Read	56	0.2%
macvendor	Read	56	0.2%
paste_email	Read	56	0.2%
robotsmeta	Read	56	0.2%
vigenere	Read	56	0.2%
bssid_geo	Read	55	0.2%
cpe	Read	55	0.2%
cve_poc_checker	Read	55	0.2%
emailrep	Read	55	0.2%
exploit_cve	Read	55	0.2%
ipv4classify	Read	55	0.2%
trufflehog_url	Read	55	0.2%
username_dossier	Read	55	0.2%
cryptomarket	Read	54	0.2%
cve_timeline	Read	54	0.2%
email_registration	Read	54	0.2%
gps_deep	Read	54	0.2%
internetdb	Read	54	0.2%
latlonformat	Read	54	0.2%
numlookup	Read	54	0.2%
person_identity	Read	54	0.2%
phonenanp	Read	54	0.2%
quakes	Read	54	0.2%
smtpbanner	Read	54	0.2%
urlscan	Read	54	0.2%
api_key_scan	Read	53	0.2%
atbash	Read	53	0.2%
cidr	Read	53	0.2%
email_crosssite	Read	53	0.2%
feodoips	Read	53	0.2%
gh_dorking	Read	53	0.2%
hashtext	Read	53	0.2%
hudsonrock	Read	53	0.2%
hunter_find	Read	53	0.2%
paste_domain	Read	53	0.2%
spaceweather	Read	53	0.2%
swiftbic	Read	53	0.2%
wayback_leaks	Read	53	0.2%
whatsnearby	Read	53	0.2%
whoisserver	Read	53	0.2%
human_sweep	Execute	52	0.2%
base36	Read	52	0.2%
caesar	Read	52	0.2%
casify	Read	52	0.2%
checksum	Read	52	0.2%
cve	Read	52	0.2%
cvedetail	Read	52	0.2%
cve_severity	Read	52	0.2%
dns	Read	52	0.2%
dnsptrrange	Read	52	0.2%
ghadvisory	Read	52	0.2%
hunter_domain	Read	52	0.2%
imagephash	Read	52	0.2%
imei	Read	52	0.2%
mgrs	Read	52	0.2%
npmdownloads	Read	52	0.2%
nvdcve	Read	52	0.2%
onthisday	Read	52	0.2%
osv	Read	52	0.2%
phonefmt	Read	52	0.2%
rdap	Read	52	0.2%
ripewhois	Read	52	0.2%
ssn_check	Read	52	0.2%
teamcymru	Read	52	0.2%
unixperm	Read	52	0.2%
utm	Read	52	0.2%
asnprefixes	Read	51	0.2%
cdxwayback	Read	51	0.2%
clickjacking	Read	51	0.2%
cratestats	Read	51	0.2%
cryptoprice	Read	51	0.2%
deviantart	Read	51	0.2%
dnsbl	Read	51	0.2%
ean	Read	51	0.2%
feodo	Read	51	0.2%
ghevents	Read	51	0.2%
gh_secret_scan	Read	51	0.2%
gleif_name	Read	51	0.2%
goproxy	Read	51	0.2%
hexdump	Read	51	0.2%
homoglyph	Read	51	0.2%
hunter_verify	Read	51	0.2%
imagehash	Read	51	0.2%
ip_math	Read	51	0.2%
leakcheck	Read	51	0.2%
passwordcheck	Read	51	0.2%
phone_leak	Read	51	0.2%
phonepivot	Read	51	0.2%
revimg	Read	51	0.2%
s3buckets	Read	51	0.2%
subbrute	Read	51	0.2%
timezone_info	Read	51	0.2%
wordcount	Read	51	0.2%
dockerhub	Write	51	0.2%
imagexmp	Write	51	0.2%
apileak	Read	50	0.2%
arxiv	Read	50	0.2%
camera	Read	50	0.2%
cisa_alerts	Read	50	0.2%
coininfo	Read	50	0.2%
dehashed_domain	Read	50	0.2%
dnssec	Read	50	0.2%
dnsverify	Read	50	0.2%
email_domain_cross	Read	50	0.2%
flightsnear	Read	50	0.2%
gdelt	Read	50	0.2%
ghkeysgpg	Read	50	0.2%
gitexposed	Read	50	0.2%
github_code	Read	50	0.2%
greynoise	Read	50	0.2%
hibp_breaches	Read	50	0.2%
hstspreload	Read	50	0.2%
hunter_count	Read	50	0.2%
imageexif	Read	50	0.2%
imagelsb	Read	50	0.2%
imagerev	Read	50	0.2%
ipint	Read	50	0.2%
maidenhead	Read	50	0.2%
morse	Read	50	0.2%
nato	Read	50	0.2%
peeringnet	Read	50	0.2%
pluscode	Read	50	0.2%
regdomain	Read	50	0.2%
restcountry	Read	50	0.2%
revgeo	Read	50	0.2%
roman	Read	50	0.2%
sec_edgar	Read	50	0.2%
suntimes	Read	50	0.2%
tempmail	Read	50	0.2%
tiktok	Read	50	0.2%
tldinfo	Read	50	0.2%
unicode_lookup	Read	50	0.2%
wikipedia	Read	50	0.2%
zerodday_timeline	Read	50	0.2%
abusecontact	Read	49	0.2%
antipode	Read	49	0.2%
barcode	Read	49	0.2%
base32	Read	49	0.2%
bin	Read	49	0.2%
btcfees	Read	49	0.2%
color	Read	49	0.2%
comb_search	Read	49	0.2%
creditcardtest	Read	49	0.2%
dnsprop	Read	49	0.2%
geohash	Read	49	0.2%
hnuser	Read	49	0.2%
hosthunt	Read	49	0.2%
iban	Read	49	0.2%
imagegps	Read	49	0.2%
ipwhois	Read	49	0.2%
jslibs	Read	49	0.2%
kev	Read	49	0.2%
luhn	Read	49	0.2%
mitre_technique	Read	49	0.2%
nstrace	Read	49	0.2%
otxdomain	Read	49	0.2%
otxip	Read	49	0.2%
password_dumps	Read	49	0.2%
peeringdb	Read	49	0.2%
railfence	Read	49	0.2%
randomuser	Read	49	0.2%
ransomware_tracker	Read	49	0.2%
snowflake	Read	49	0.2%
spacepeople	Read	49	0.2%
srvlookup	Read	49	0.2%
supply_chain_risk	Read	49	0.2%
swagger	Read	49	0.2%
tlsa	Read	49	0.2%
vin	Read	49	0.2%
whois	Read	49	0.2%
wikidata	Read	49	0.2%
wpscan	Read	49	0.2%
zonetransfer	Read	49	0.2%
base_convert	Write	49	0.2%
formaudit	Write	49	0.2%
csp_parse	Execute	48	0.2%
transform	Execute	48	0.2%
agify	Read	48	0.2%
asnlookup	Read	48	0.2%
botnet_tracker	Read	48	0.2%
breachdirectory	Read	48	0.2%
cidrinfo	Read	48	0.2%
circlhash	Read	48	0.2%
crossrefauthor	Read	48	0.2%
ddg_instant	Read	48	0.2%
dnsmx	Read	48	0.2%
genderize	Read	48	0.2%
geocode	Read	48	0.2%
geodist	Read	48	0.2%
ghpubkeys	Read	48	0.2%
ghrepo	Read	48	0.2%
graphql	Read	48	0.2%
huggingface	Read	48	0.2%
imagethumb	Read	48	0.2%
ipfull	Read	48	0.2%
ipgeo	Read	48	0.2%
jwtdecode	Read	48	0.2%
lei	Read	48	0.2%
macvendorlookup	Read	48	0.2%
malware_family	Read	48	0.2%
marineweather	Read	48	0.2%
maven	Read	48	0.2%
musicbrainz	Read	48	0.2%
openalexwork	Read	48	0.2%
openfoodfacts	Read	48	0.2%
orcid	Read	48	0.2%
orgname	Read	48	0.2%
packagist	Read	48	0.2%
phoneapps	Read	48	0.2%
phonevcard	Read	48	0.2%
reddit_sub	Read	48	0.2%
referrer_pol	Read	48	0.2%
robotstxt_history	Read	48	0.2%
rot47	Read	48	0.2%
rpki	Read	48	0.2%
spdxlicense	Read	48	0.2%
sri_check	Read	48	0.2%
ulid	Read	48	0.2%
unpaywall	Read	48	0.2%
urlscansearch	Read	48	0.2%
uuid	Read	48	0.2%
vimeo	Read	48	0.2%
waf	Read	48	0.2%
wikidatasearch	Read	48	0.2%
youtube	Read	48	0.2%
apod	Read	47	0.2%
asrank	Read	47	0.2%
bgphistory	Read	47	0.2%
binarytext	Read	47	0.2%
binlookup	Read	47	0.2%
caa	Read	47	0.2%
cfradar	Read	47	0.2%
cloud	Read	47	0.2%
creditcard	Read	47	0.2%
ctlogsearch	Read	47	0.2%
darkweb_monitor	Read	47	0.2%
datacite	Read	47	0.2%
decode	Read	47	0.2%
epss	Read	47	0.2%
gitignore	Read	47	0.2%
gravatarfull	Read	47	0.2%
holidays	Read	47	0.2%
htmlcomments	Read	47	0.2%
httpcode	Read	47	0.2%
imagemeta	Read	47	0.2%
isbnmeta	Read	47	0.2%
isin	Read	47	0.2%
jsonfmt	Read	47	0.2%
leetcode_user	Read	47	0.2%
links	Read	47	0.2%
metatags	Read	47	0.2%
musicbrainzartist	Read	47	0.2%
openalexauthor	Read	47	0.2%
orcidworks	Read	47	0.2%
passentropy	Read	47	0.2%
passgen	Read	47	0.2%
permissions_pol	Read	47	0.2%
pgp	Read	47	0.2%
phishing_intel	Read	47	0.2%
phonecc	Read	47	0.2%
rdapip	Read	47	0.2%
reddit_user	Read	47	0.2%
revgeocode	Read	47	0.2%
rubygemrev	Read	47	0.2%
sha256lookup	Read	47	0.2%
sopostuser	Read	47	0.2%
soundcloud	Read	47	0.2%
tempconv	Read	47	0.2%
tls	Read	47	0.2%
typosquat	Read	47	0.2%
useragent	Read	47	0.2%
webfinger	Read	47	0.2%
whois_check	Read	47	0.2%
apt_track	Read	46	0.2%
ascii85	Read	46	0.2%
attack_surface	Read	46	0.2%
base64	Read	46	0.2%
bimi	Read	46	0.2%
breachsearch	Read	46	0.2%
c2_infrastructure	Read	46	0.2%
commoncrawl	Read	46	0.2%
cpfcnpj	Read	46	0.2%
cratedownloads	Read	46	0.2%
crc32	Read	46	0.2%
credential_stuffing	Read	46	0.2%
cron	Read	46	0.2%
dnsquery	Read	46	0.2%
dogeaddr	Read	46	0.2%
elevation	Read	46	0.2%
emoji	Read	46	0.2%
entropy	Read	46	0.2%
favicon	Read	46	0.2%
feeds	Read	46	0.2%
ghcommits	Read	46	0.2%
ghgists	Read	46	0.2%
github_trending	Read	46	0.2%
hackernews	Read	46	0.2%
hackernews_front	Read	46	0.2%
httpmethods	Read	46	0.2%
ipv6	Read	46	0.2%
isexitnode	Read	46	0.2%
isotime	Read	46	0.2%
linktree	Read	46	0.2%
macvalid	Read	46	0.2%
manifest	Read	46	0.2%
mimetype	Read	46	0.2%
nationalize	Read	46	0.2%
npmpkg	Read	46	0.2%
phonespam	Read	46	0.2%
portlookup	Read	46	0.2%
portquick	Read	46	0.2%
pubmed	Read	46	0.2%
pypiproject	Read	46	0.2%
quoted_printable	Read	46	0.2%
semanticscholar	Read	46	0.2%
spamhauslookup	Read	46	0.2%
takeover	Read	46	0.2%
tlsscan	Read	46	0.2%
txhash	Read	46	0.2%
uuid_validate	Read	46	0.2%
vatid	Read	46	0.2%
weather	Read	46	0.2%
wellknown	Read	46	0.2%
osmuser	Write	46	0.2%
adstxt	Read	45	0.2%
asn	Read	45	0.2%
bearing	Read	45	0.2%
breachdb	Read	45	0.2%
btcaddr	Read	45	0.2%
cookies	Read	45	0.2%
dirlisting	Read	45	0.2%
disposable	Read	45	0.2%
disposablecheck	Read	45	0.2%
dnsrecon	Read	45	0.2%
domainage	Read	45	0.2%
emailformat	Read	45	0.2%
emailsec	Read	45	0.2%
ethcontract	Read	45	0.2%
ghlanguages	Read	45	0.2%
githubgists	Read	45	0.2%
githubsearch	Read	45	0.2%
gomod	Read	45	0.2%
hostsearch	Read	45	0.2%
httpstatus	Read	45	0.2%
ioc_reputation	Read	45	0.2%
ltcaddr	Read	45	0.2%
mastodon	Read	45	0.2%
moonphase	Read	45	0.2%
mtasts	Read	45	0.2%
numwords	Read	45	0.2%
openphish	Read	45	0.2%
peers	Read	45	0.2%
phone	Read	45	0.2%
portscan	Read	45	0.2%
pypistats	Read	45	0.2%
qrcode	Read	45	0.2%
rdap_domain	Read	45	0.2%
rot13	Read	45	0.2%
securitytxt	Read	45	0.2%
threatcrowd	Read	45	0.2%
tor	Read	45	0.2%
weatheralerts	Read	45	0.2%
wikisummary	Read	45	0.2%
xrpaddr	Read	45	0.2%
datauri	Write	45	0.2%
wikipv	Write	45	0.2%
wpplugin	Write	45	0.2%
brew	Read	44	0.2%
certhistory	Read	44	0.2%
codeberg	Read	44	0.2%
codeforces	Read	44	0.2%
cors	Read	44	0.2%
depsdev	Read	44	0.2%
dnsgraph	Read	44	0.2%
doi	Read	44	0.2%
dorks	Read	44	0.2%
ens	Read	44	0.2%
ethaddr	Read	44	0.2%
fxrate	Read	44	0.2%
golangpkg	Read	44	0.2%
hashid	Read	44	0.2%
hexpm	Read	44	0.2%
htmlencode	Read	44	0.2%
imageicc	Read	44	0.2%
infra_fingerprint	Read	44	0.2%
isbn	Read	44	0.2%
jwt	Read	44	0.2%
leet	Read	44	0.2%
lobsters	Read	44	0.2%
port	Read	44	0.2%
pypi	Read	44	0.2%
rubygems	Read	44	0.2%
slug	Read	44	0.2%
soladdr	Read	44	0.2%
stackoverflow	Read	44	0.2%
telegram_channel	Read	44	0.2%
threat_actor	Read	44	0.2%
urlparse	Read	44	0.2%
zenodo	Read	44	0.2%
blockheight	Read	43	0.2%
bluesky	Read	43	0.2%
chesscom	Read	43	0.2%
country	Read	43	0.2%
crates	Read	43	0.2%
crypto	Read	43	0.2%
devto	Read	43	0.2%
email	Read	43	0.2%
hashnode	Read	43	0.2%
headers	Read	43	0.2%
iss	Read	43	0.2%
keybase	Read	43	0.2%
nearbywiki	Read	43	0.2%
npmdl	Read	43	0.2%
npmorg	Read	43	0.2%
nuget	Read	43	0.2%
opencorp	Read	43	0.2%
pageinfo	Read	43	0.2%
robots	Read	43	0.2%
semver	Read	43	0.2%
sitemap	Read	43	0.2%
social	Read	43	0.2%
ssh	Read	43	0.2%
steam_user	Read	43	0.2%
stringmetrics	Read	43	0.2%
subdomains	Read	43	0.2%
threat_pattern	Read	43	0.2%
trendingwiki	Read	43	0.2%
urban	Read	43	0.2%
archiveorg	Read	42	0.2%
dblp	Read	42	0.2%
epoch	Read	42	0.2%
github	Read	42	0.2%
hnsearch	Read	42	0.2%
hostname	Read	42	0.2%
httping	Read	42	0.2%
imagecolors	Read	42	0.2%
pwstrength	Read	42	0.2%
redirects	Read	42	0.2%
reversedns	Read	42	0.2%
reverseip	Read	42	0.2%
username	Read	42	0.2%
uuencode	Read	42	0.2%
wayback	Read	42	0.2%
cdnjs	Read	41	0.2%
fccid	Read	41	0.2%
ghorg	Read	41	0.2%
mac	Read	41	0.2%
medium	Read	41	0.2%
npm	Read	41	0.2%
tech	Read	41	0.2%
gitlab	Read	40	0.2%
random_fact	Read	40	0.2%
agent_manifest	Read	36	0.2%

// CUT THE COST

Your agent uses a handful of these tools. It pays for all 474.

You don't need all 474 of those definitions in the window. PolicyLayer is an MCP gateway that sits in front of Osint Terminal: only the tools you grant are exposed to the agent, the rest never load. A smaller window means a sharper agent — less noise when it picks a tool — and every request costs less:

Grant scope	Definition cost	Reduction
All 474 tools (no gateway)	22,880 tokens	—
3 granted tools	~145 tokens	−99%
5 granted tools	~241 tokens	−99%
10 granted tools	~483 tokens	−98%

Create a free account and register Osint Terminal — nothing to install.
Grant only the tools you use — ungranted definitions never enter the context window.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

CUT OSINT TERMINAL TOKEN COST →

Instant setup, no code required.

Model your own stack in the token-cost calculator, or see the Osint Terminal policy for what a sensible grant looks like.

//FAQ

Osint Terminal token-cost questions.

How many tokens does the Osint Terminal MCP server use?+

Its 474 tool definitions total 22,880 tokens — 11% of a 200k context window — measured with tiktoken o200k_base over the serialised tools/list payload. Exact counts vary slightly by client and model.

Why does Osint Terminal consume tokens before I send a message?+

MCP clients load every connected server's tool definitions — name, description, and input schema — into the model's context so it knows what it can call. That payload is charged against your context window on every request, whether or not a tool is used.

How do I reduce Osint Terminal's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Osint Terminal to only the tools you allow — ungranted definitions are filtered out of the tool list, so they never enter the context window. A grant of 3 typical tools costs roughly 145 tokens, a 99% reduction.

Does deferred tool loading fix this?+

Partially, in some clients. Claude Code defers MCP tool schemas behind a tool-search step by default, and VS Code has experimental grouping — but you still pay tokens per search and reload, and Cursor, Windsurf and Gemini CLI load definitions upfront. Reducing the exposed tool set cuts the cost in every client.

// METHOD & REVIEW

How these numbers were measured.

01

Serialisation

Each tool is serialised as a tools/list entry — name, description, input schema — from the schemas in the PolicyLayer scan database. Clients differ slightly in framing, so treat counts as close estimates.

02

Tokeniser

tiktoken o200k_base (GPT-4o/o-series). Anthropic's current tokeniser isn't published, so Claude's exact counts will differ; for English text and JSON schemas the totals are close enough to treat these as estimates.

03

Deferred loading

Some clients now defer schema loading (Claude Code's tool search; VS Code experimental grouping). You still pay per search and reload — and Cursor, Windsurf and Gemini CLI load everything upfront.

Computed 20-10-2026 from the PolicyLayer scan database over all 474 catalogued Osint Terminal tools. Counts refresh with every site build.