get_species and explain_item inline base64 images that blow the MCP response budget (286 KB for a 150-char wiki extract) #230
Labels
No labels
burndown-2026-06
headless
autonomy
async-consult
autonomy
epic
autonomy
headless
autonomy
live-collab
c#
priority
P0
priority
P1
priority
P2
priority
P3
priority
P4
role/ai
role/creator
role/design
role/director
role/engineer
role/exec
role/human
role/ops
role/qa
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
coilyco-gaming/eco-app#230
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
QA sweep, 2026-08-12.
Observed
get_species(name="Snapping Turtle")returns 286,264 characters. Field sizes:photoDataUriwikiExtracttaxonomyerror99.8% of the response is one base64 JPEG. The response exceeds the MCP client's token cap, so a caller gets an error and a file path instead of an answer — for a lookup whose actual content is two sentences and a taxon.
explain_item(name="Wheat")has the same shape: ~100 KB ofimage_data_uriaround a 3-sentence description.Expected
Serve the image by URL and let the client fetch it, or gate it behind an opt-in parameter (
include_image: falseby default). A tool whose useful payload is 200 bytes should not cost a quarter of a megabyte.Context
7 of 22 tools exceeded the response cap in this sweep. This is the most extreme case and the easiest fix.