Add error handling on the calls to the server to debug issue #132 - #137
Merged
Merged
Conversation
Closed
Member
|
hey @tmeire sorry, just seeing this and i reworked init quite a bit! want to take a look at latest and see if your changes still apply? |
Contributor
Author
|
@adamdottv merged the changes from dev and moved the http status code checks to the new locations 👍 |
Member
|
really sorry, things moved again, mind to rebase? also, we have toast messages now if you'd like to notify the user that way |
tmeire
force-pushed
the
fix/server-error-reporting
branch
from
June 21, 2025 07:38
2c4ef78 to
1294118
Compare
Contributor
Author
|
@adamdotdevin rebased again If it bails out on the providers cmd, it's still crashing because multiple parts in the rest of the code assume there's always a provider configured. |
achembarpu
pushed a commit
to achembarpu/opencode
that referenced
this pull request
Aug 4, 2025
randomm
referenced
this pull request
in randomm/opencode
Jan 9, 2026
xywsxp
pushed a commit
to xywsxp/opencode
that referenced
this pull request
Apr 24, 2026
bussard76
pushed a commit
to bussard76/openwork
that referenced
this pull request
May 12, 2026
bussard76
pushed a commit
to bussard76/openwork
that referenced
this pull request
May 12, 2026
…#137) (anomalyco#437)" This reverts commit f1d336d.
AIALRA-0
pushed a commit
to AIALRA-0/opencode-turn-engine
that referenced
this pull request
Jun 10, 2026
avion23
pushed a commit
to avion23/opencode
that referenced
this pull request
Jun 10, 2026
6 tasks
mayoalexander
added a commit
to FREELABEL/iris-opencode
that referenced
this pull request
Aug 18, 2026
Same model, same question, six hours apart:
17:01 'fetch my family projects' -> bloqs list, then bloqs SEARCH family,
found anomalyco#200 Family Businesses and anomalyco#137 Family Health and Finance
Management Workflow. Correct.
21:58 'what personal + family project do i have' -> bloqs list ONLY, answered
'Mia Mayo — Life Atlas' and stopped. Missed both.
Neither anomalyco#200 nor anomalyco#137 appears anywhere in 'bloqs list' — it is recency-ordered and
truncated. So the list-only answer is not partial, it is WRONG, and nothing in the
result tells the model that. The list looks complete.
kimi-k3 and glm-5.2 failed the same way on the same prompt, both stopping after one
list call and guessing from names. That is three models making one mistake, which
makes it a tool-description problem rather than a model problem.
The description now states plainly that 'bloqs list' is not an inventory, gives the
measured example, and says: search by topic, list only to enumerate recent boards or
when the user already named one.
Also covers the second failure in that transcript. Asked 'are there any other
projects?', the model ran 'bloqs get 602 --items' — which returns what is INSIDE 602
and cannot say anything about other boards — then answered from memory of the earlier
list. Both are now called out explicitly, along with a reminder that an oversized
result saved to a file is a success.
Put in the TOOL DESCRIPTION rather than a prompt file deliberately: it reaches every
model through the same channel, and the default model has changed three times this
week.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
mayoalexander
added a commit
to FREELABEL/iris-opencode
that referenced
this pull request
Aug 18, 2026
…ot everything Two data-contract defects, both measured, both of which read as model failures and were not. Written test-first; the tests fail against the old behaviour. 1. TRUNCATION WAS INVISIBLE. `bloqs list --json` returned 20 records as a FLAT ARRAY. The account has 137. No total, no truncated flag, no next — so a model reading it reasonably concluded it had everything, then answered "your family project is anomalyco#602" while anomalyco#200 Family Businesses, anomalyco#137 Family Health and Finance, anomalyco#544 and anomalyco#584 sat in the other 117. The API sends the total; the CLI dropped it on the floor. Now: {data, meta:{returned, total, truncated, limit, hint}}. Verified live — returned 20, total 137, truncated true. When the server sends no total, a FULL page is treated as truncated. Claiming "complete" on no evidence is exactly the failure being fixed: over-reporting costs one extra query, under-reporting produces a confident wrong answer. 2. `list` HAD NO FIELD CONTRACT. bloqs list = 20 records x 4 fields = 2,290 bytes agents list = 30 records x 72 fields = 145,395 bytes Same verb, 63x apart. The second overflowed the MCP every time, forcing a spill -> jq -> narrow round-trip to answer "what agents do I have". LIST_FIELDS caps a listing at 8 fields — identity plus discriminators — and a test enforces the ceiling plus the presence of an id and a human-readable label, because a row you cannot follow up with `get` is not lean, it is useless. `agents get <id>` still returns the full record. Measured: agents list 145,395 -> 7,828 bytes. 18.6x. It no longer overflows. WHY THIS RATHER THAN MORE PROMPTING. I previously diagnosed the missed-bloqs bug as a prompt problem and wrote tool-description guidance for it. That guidance still helps, but it was teaching models to compensate for a payload that misleads them. A tool returning 15% of the data while looking complete will defeat any model; one returning 145KB for a listing makes an easy question expensive. Fix the contract, not the reader. Shape change on --json for these two commands: consumers doing `jq '.[]'` now need `jq '.data[]'`. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds HTTP response code checking on a couple of calls to the server.