Mutual observation

The observatory reads itself being read.

Civic systems get read through their records. The agent surface gets read through its invocation log. Every call to the public MCP server at /mcp and /api/mcp/{tool} lands in an append-only table; aggregates surface here. Misuses, missed capabilities, and quiet successes — they all count. This page is the public reading of how AI agents read the field.

Trailing 30 days

27K invocations · 651 sessions

Total invocations27K27K ok · 69 error

Success rate99.7%Median 0 ms

Unique sessions6514 agent families

Tools called17of the published catalog

Top tools

What agents asked the field

Tool	Invocations	Success	Median ms
`list_corridors`	1.3K	100%	3
`list_places`	455	100%	13
`submit_agent_feedback`	394	99%	59
`semantic_search`	319	100%	283
`describe_zoning_signal`	160	100%	0
`describe_corridor`	55	54.5%	208
`list_entities`	37	86.5%	25
`meeting_index`	34	100%	216
`describe_place`	26	80.8%	248
`list_patterns`	12	100%	36
`list_watch_items`	12	100%	20
`describe_entity`	10	0%	2
`get_track_record`	7	100%	35
`list_meetings`	5	100%	289
`describe_pattern`	4	25%	1
`__verifymcp_auth_probe_9fbd979945ca67bb__`	1	0%	0
`__verifymcp_auth_probe_f972be8d4566858d__`	1	0%	0

Agent families

Who is reading the field

Family	Invocations	Sessions	Success rate
unknown	27K	651	99.8%
browser	150	0	98.7%
anthropic-sdk	6	0	100%
curl	5	0	100%

Most-read slugs

Which corner of the field gets asked about

Slug	Via	Invocations
`sr-429-western-beltway`	`describe_corridor`	25
`clermont-florida`	`describe_place`	8
`apopka-florida`	`describe_place`	7
`Leesburg`	`meeting_index`	6
`denver`	`describe_corridor`	5
`texas-triangle`	`describe_corridor`	5
`us-27-south-lake`	`describe_corridor`	5
`i-35-corridor`	`describe_corridor`	5
`Clermont`	`meeting_index`	5
`gulf-coast`	`describe_corridor`	5
`piedmont-atlantic-corridor`	`describe_entity`	5
`massachusetts`	`describe_corridor`	5
`piedmont`	`describe_entity`	5
`san-diego-california`	`describe_place`	5
`Seattle`	`meeting_index`	4

Feedback

390 submissions — what agents told us

Category	Submitted	Reviewed
praise	312	0
gap	42	0
suggestion	27	0
error	9	0

Methodology

How the readings are produced

Every tool invocation against the public MCP surface writes one row to zoning_signal_mcp_invocations with timestamp, transport, tool name, sanitized arguments, result status, duration, derived agent family, and a salt-hashed IP. Raw IPs are never stored. Per-row reads are RLS-denied; only aggregate RPCs reach this surface, granted to anonymous callers. Aggregates refresh on a minutes-scale cache. The feedback table mirrors the same privacy posture — anyone may submit, only operators may read, only the operator-curated surfaced subset appears here.