Atlas entry

`overwing.ai`

43 capabilities over MCP. The inventory is incomplete. First atlas observation.

Observed 12 Nov 2026, 02:05 UTC · passive: no tools called, no credentials sent · sha256:436b7e…2983

43 tools. 19 only read. 2 only compute. 7 change records, and 2 of those delete or cancel. 1 sends messages. 14 have no reading.

AI reading, beta. side-effects v1 on claude-haiku-5.5, calibrated on MCP (golden-3). Only readings above a measured precision count. Counts are tools, not calls: no tool was called.

AI readings summary · beta

The service lists 43 tools, and 29 have AI readings.

atlas_withdraw_registration and tower_revoke_agent are read as deleting or cancelling records.

feedback is read as sending messages to people.

5 tools are read as creating records, including create_rule_set, cad_create and atlas_register_agent.

Written by code from the AI readings on this page: it groups them and says nothing they do not.

What an agent can do here

AI readings, beta · Readings on 29 of 43 MCP tools. 2 read as deleting or cancelling, 1 reads as sending messages.

Moves money
0
No tool read this way.
Deletes or cancels records
2 tools
- atlas_withdraw_registration0.88
- tower_revoke_agent0.84
Sends messages to people
1 tool
- feedback0.84
Changes permissions
0
No tool read this way.
Runs code
0
No tool read this way.
Should ask a person first
0
No tool read this way.

No counted reading 14 tools No reading was counted, so any effect above is possible for them: atlas_lookup, atlas_verify_registration, beacon_report, beacon_start, evaluate, evaluate_batch, preflight_check, preflight_payment, preflight_report, prove_domain and 4 more.

What changed

First atlas observation. Atlas history

Look first

AI reading, beta: AI adds items here; it never removes one.

atlas_withdraw_registrationDeletes or cancels records · 0.88
tower_revoke_agentDeletes or cancels records · 0.84
feedbackSends messages to people · 0.84

Surfaces

5 protocols checked

ARD	Not found	No document retrieved
A2A	invalid	0 skills
MCP	Valid	2026-07-28 · 43 tools · resources not advertised · prompts not advertised
OpenAPI	Not detected	0 operations
WebMCP	Not detected	0 declarative form tools · experimental

Exact states and meaning

ARD
absent
Exact state: not found · No document retrieved
The supported discovery location returned no matching document.
Not found does not prove the service exposes no such interface elsewhere.
Not observed / not fetched
A2A
failed
Exact state: invalid · Revision 1.0 · 0 skills
InterfaceDelta retained the machine state “invalid” without assigning an additional meaning.
Not observed / not fetched
MCP
observed
Exact state: valid · Revision 2026-07-28 · 43 tools · resources not advertised · prompts not advertised
A usable protocol document was retrieved and validated.
Directly observed
OpenAPI
absent
Exact state: not detected · 0 operations
No endpoint was established from the evidence and scope that InterfaceDelta inspected.
Not detected does not prove that no endpoint exists.
Not observed / not fetched
WebMCP
absent
Exact state: not detected · 0 declarative form tools · experimental
No named declarative WebMCP form tools were observed in the origin root HTML.
Script-registered tools and other pages are not observed, so this does not mean the site has no WebMCP tools.
Not observed / not fetched

43 · inventory incomplete

Showing 43 of 43.

Tool and inputs							◇ AI reading, beta · confidence
Read as one of the six effects 3 AI only raises attention: these rows come first.
id*◇ AIDeletes or cancels records 0.88		Deletes or cancels records 0.88					Deletes or cancels records0.88
`atlas_withdraw_registration` inputs `id` (required) AI Deletes or cancels records 0.88
about, contact, evaluation_id, message*◇ AISends messages to people 0.84			Sends messages to people 0.84				Sends messages to people0.84
`feedback` inputs `about`, `contact`, `evaluation_id`, `message` (required) AI Sends messages to people 0.84
agent_id*◇ AIDeletes or cancels records 0.84		Deletes or cancels records 0.84					Deletes or cancels records0.84
`tower_revoke_agent` inputs `agent_id` (required) AI Deletes or cancels records 0.84
Read as none of the six 26 Not a clearance.
limit, operator, purpose, q and 1 more							Reads only0.86
`atlas_agents` inputs `limit`, `operator`, `purpose`, `q` and 1 more AI Reads only 0.86
no inputs							Reads only0.88
`atlas_list_registrations` AI Reads only 0.88
description, domain*, follows_robots_txt, key_directory_url and 6 more							Creates records0.90
`atlas_register_agent` inputs `description`, `domain` (required), `follows_robots_txt`, `key_directory_url` and 6 more AI Creates records 0.90
no inputs							Reads only0.85
`atlas_summary` AI Reads only 0.85
no inputs							Reads only0.88
`beacon_sample` AI Reads only 0.88
from, idempotency_key, params, prompt*							Creates records0.80
`cad_create` inputs `from`, `idempotency_key`, `params`, `prompt` (required) AI Creates records 0.80
id							Reads only0.86
`cad_get` inputs `id` AI Reads only 0.86
org_name, use_for_session							Creates records0.86
`create_account` inputs `org_name`, `use_for_session` AI Creates records 0.86
description, name, rules, slug*							Creates records0.90
`create_rule_set` inputs `description`, `name` (required), `rules` (required), `slug` (required) AI Creates records 0.90
id*							Reads only0.91
`get_evaluation` inputs `id` (required) AI Reads only 0.91
slug*							Reads only0.88
`get_rule_set` inputs `slug` (required) AI Reads only 0.88
days							Reads only0.85
`get_usage` inputs `days` AI Reads only 0.85
cursor, limit, rule_set, verdict							Reads only0.90
`list_evaluations` inputs `cursor`, `limit`, `rule_set`, `verdict` AI Reads only 0.90
no inputs							Reads only0.90
`list_plans` AI Reads only 0.90
include_inactive							Reads only0.90
`list_rule_sets` inputs `include_inactive` AI Reads only 0.90
no inputs							Computes only, changes nothing0.80
`preflight_overview` AI Computes only, changes nothing 0.80
no inputs							Reads only0.85
`preflight_record` AI Reads only 0.85
id*							Reads only0.90
`preflight_verdict` inputs `id` (required) AI Reads only 0.90
no inputs							Reads only0.84
`tower_capabilities` AI Reads only 0.84
name, scopes, use_for_session							Creates records0.84
`tower_create_agent` inputs `name` (required), `scopes` (required), `use_for_session` AI Creates records 0.84
input, operation							Computes only, changes nothing0.80
`tower_decide` inputs `input` (required), `operation` (required) AI Computes only, changes nothing 0.80
action_id*							Reads only0.85
`tower_get_action` inputs `action_id` (required) AI Reads only 0.85
id*							Reads only0.90
`tower_get_receipt` inputs `id` (required) AI Reads only 0.90
no inputs							Reads only0.86
`tower_list_agents` AI Reads only 0.86
from, to							Reads only0.85
`tower_verify_receipts` inputs `from`, `to` AI Reads only 0.85
no inputs							Reads only0.85
`whoami` AI Reads only 0.85
No counted reading 14 No reading was counted, so any effect is possible.
user_agent*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`atlas_lookup` inputs `user_agent` (required)
id*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`atlas_verify_registration` inputs `id` (required)
id*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`beacon_report` inputs `id` (required)
url*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`beacon_start` inputs `url` (required)
context, input*, metadata, rule_set and 1 more	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`evaluate` inputs `context`, `input` (required), `metadata`, `rule_set` and 1 more
context, items*, rule_set, store	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`evaluate_batch` inputs `context`, `items` (required), `rule_set`, `store`
policy, transaction	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`preflight_check` inputs `policy` (required), `transaction` (required)
network, payment_required*, policy, requested_url	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`preflight_payment` inputs `network`, `payment_required` (required), `policy`, `requested_url`
id, signature	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`preflight_report` inputs `id` (required), `signature` (required)
domain*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`prove_domain` inputs `domain` (required)
action_id*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`tower_compensate` inputs `action_id` (required)
no inputs	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`tower_load_template`
dry_run, idempotency_key, input, operation*	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`tower_submit_action` inputs `dry_run`, `idempotency_key` (required), `input` (required), `operation` (required)
no inputs	unknown	unknown	unknown	unknown	unknown	unknown	No reading
`verify_domain`

Read as this effect No reading: unknown Not read as this effect◇ AI readings, beta. A zero is a count, not a clearance.

A2A skills 0+ observed

Inventory incomplete. The displayed count is a lower bound; missing items do not prove removal.

None recorded in this Passport.

Not observed in this Passport: MCP resources, MCP prompts, OpenAPI operations.

The atlas shows each capability by its plain identifier, inputs and declared sign-in. Publisher descriptions, paths, links and source evidence are not published here.

The comparison with the previous observation is shown only to the account that recorded it.

Atlas history

1 observation

12 Nov 2026, 02:05 UTC43 capabilities · MCP 2026-07-28 · MCP endpoint validFirst observation
Fingerprint and versions
Fingerprint
436b7edb2ac3529738dfb3280cb2bf8fa60a55b459a320ccd7b977a442362983
Passport schema
0.5.0, passport-semantic-v4
Consistency
MCP_PROMPTS_ADVERTISEMENT_LIST: not applicable · MCP_RESOURCES_ADVERTISEMENT_LIST: not applicable · MCP_TOOLS_ADVERTISEMENT_LIST: match · OPENAPI_MCP_EXPLICIT_MAPPING: not applicable

Each class compares an observation with the previous atlas observation. Runs of unchanged observations are grouped. Struck values are what the previous observation recorded; inserted values are what this one recorded.

Declared and observed

43 tools declare hints; the AI readings agree with all 29 that have a reading.

A disagreement is not a finding about the service: either side can be wrong. How we compare declarations and readings.

About this entry

ai.overwing/mcp
Listed
Official MCP Registry · first listed 5 Oct 2026 · last listed 25 Oct 2026

Passive observations of a public interface listed in the Official MCP Registry. An observation does not verify security, ownership or compliance, and InterfaceDelta assigns no scores or rankings. Methodology · Report an inaccuracy · Ask for removal

Identity and snapshot

Target key: sha256:0fd0ae793b6abd524b4610044bf58267119698a41e9319640e42610810c2f15f
Locator SHA-256: 3e85fafb3c320f8624a5e48438c31e87cb1792c388fc5bf2326614cd972faaa0
Snapshot generated: 12 Nov 2026, 02:26 UTC

The entry is one exact observed target, not a real-world service identity, and implies no ownership. Endpoint paths are not published; the locator hash lets an operator confirm their own URL.

Capability facts

As observed from outside, at the time below

Observed: 12 Nov 2026, 02:05 UTC
Signed: 12 Nov 2026, 02:26 UTC

Protocols: MCP
Tools: 43
Declared sign-in: None declared

Effects counted AI readings, beta

Moves money: 0
Deletes or cancels records: 2
Sends messages to people: 1
Changes permissions: 0
Runs code: 0
Asks a person: 0

43 tools examined; 28 counted above a measured threshold.

Declared vs read disagreements

Declared read-only, read as writing: 0
Declared destructive, read as not writing: 0
Declared non-destructive, read as deleting: 0
Declared closed-world, read as reaching out: 0

Change since previous: First observation
Label status: Current