hermes - ✅(Solved) Fix Wrapped API errors lose nested body details and misclassify transient 402s as billing failures [4 pull requests, 1 participants]

hermes2026-04-22 23:21:49

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

NousResearch/hermes-agent#14195•Fetched 2026-04-23 07:46:12

View on GitHub

Comments

Participants

Timeline

Reactions

Author

NewTurn2017

Participants

NewTurn2017

Timeline (top)

cross-referenced ×4labeled ×3referenced ×3

classify_error() walks __cause__ / __context__ to extract a nested status code, but _extract_error_body() only inspects the top-level exception. When an SDK/API error is wrapped, Hermes can keep the nested 402 status code but lose the nested body message that distinguishes transient rate limits from billing failures.

Error Message

status_code = _extract_status_code(error) walks the cause chain
body = _extract_error_body(error) does not walk the cause chain So a wrapped error like this:
outer exception: Exception("outer")
__cause__: provider/SDK exception with status_code=402 and body message "Usage limit reached, try again in 5 minutes" Wrap a mock API error with:
body = {"error": {"message": "Usage limit reached, try again in 5 minutes"}} inside an outer exception and pass the outer exception to classify_error().

Root Cause

Expected

Classify as a transient/rate-limit condition, because the nested body explicitly says to retry later.

Fix Action

Fixed

Fixed by PR: fix(agent): preserve nested error bodies during classification (https://github.com/NousResearch/hermes-agent/pull/14219)
Fixed by PR: fix(agent): preserve nested API error bodies (https://github.com/NousResearch/hermes-agent/pull/14287)

PR fix notes

PR #14219: fix(agent): preserve nested error bodies during classification

Repository: NousResearch/hermes-agent
Author: liuhao1024
State: open | merged: False
Link: https://github.com/NousResearch/hermes-agent/pull/14219

Description (problem / solution / changelog)

Summary

walk wrapped exceptions when extracting structured API error bodies
keep nested provider messages available for message-aware classification
add regression coverage for wrapped 402 rate-limit bodies

Root cause

classify_api_error() already walks __cause__ / __context__ for status codes
_extract_error_body() only inspected the top-level exception
wrapped SDK errors therefore kept status_code=402 but lost the nested body message that says the condition is transient

Fix

make _extract_error_body() traverse the same cause/context chain as _extract_status_code()
stop at the first structured body or response.json() payload found

Regression coverage

add a helper-level test proving _extract_error_body() reads the nested body from __cause__
add an end-to-end classifier test proving a wrapped 402 with "try again" now classifies as rate_limit

Testing

scripts/run_tests.sh tests/agent/test_error_classifier.py::TestExtractErrorBody::test_from_cause_chain tests/agent/test_error_classifier.py::TestClassifyApiError::test_wrapped_402_uses_nested_body_message
scripts/run_tests.sh tests/agent/test_error_classifier.py
scripts/run_tests.sh tests/agent/test_gemini_cloudcode.py::TestGeminiHttpErrorParsing::test_status_code_flows_through_error_classifier

Closes #14195

Changed files

agent/error_classifier.py (modified, +19/-13)
tests/agent/test_error_classifier.py (modified, +25/-0)

PR #14287: fix(agent): preserve nested API error bodies

Repository: NousResearch/hermes-agent
Author: LeonSGP43
State: open | merged: False
Link: https://github.com/NousResearch/hermes-agent/pull/14287

Description (problem / solution / changelog)

Summary

make _extract_error_body walk cause/context the same way _extract_status_code already does
preserve nested SDK body messages when an outer wrapper exception hides the real API payload
add regressions for wrapped body extraction and wrapped 402 transient-limit classification

Testing

python3 -m pytest -o addopts='' tests/agent/test_error_classifier.py

Fixes #14195

Changed files

agent/error_classifier.py (modified, +19/-13)
tests/agent/test_error_classifier.py (modified, +27/-0)

PR #2: fix: apply P1 issue hardening (todo/registry/retry)

Repository: aideautomation/hermes-agent
Author: aideautomation
State: closed | merged: True
Link: https://github.com/aideautomation/hermes-agent/pull/2

Description (problem / solution / changelog)

결론

GitHub P1 오픈 이슈 4건의 핵심 재현 케이스를 테스트로 먼저 추가하고, 최소 수정으로 런타임 방어 로직을 반영했습니다.

반영 이슈

변경 요약

todo_tool 하드닝 (#14185)

todos가 문자열이면 JSON 파싱 시도
파싱 실패 시 구조화 에러 반환
list 타입 검증 추가
_validate, _dedupe_by_id에서 non-dict 입력 방어

registry dispatch 하드닝 (#14186)

_normalize_tool_name() 추가
CamelCase, _tool suffix, 구분자 드리프트를 정규화 fallback으로 처리
예: TodoTool_tool -> todo, WriteFile_tool -> write_file

run_agent 분류 수정 (#14271)

local validation fast-fail 분류에서 json.JSONDecodeError 제외
transient JSON 파싱 실패가 retry 경로를 타도록 보정

error classifier 보강 (#14195)

_extract_error_body()가 __cause__/__context__ 체인을 따라 nested body를 추출하도록 수정
wrapped 402 오류에서 nested message("usage limit ... try again")를 잃지 않도록 보정
billing 오분류를 줄이고 transient rate_limit 분류 정확도 개선

테스트

scripts/run_tests.sh tests/tools/test_todo_tool.py tests/tools/test_registry.py tests/run_agent/test_run_agent.py::TestRetryExhaustion::test_jsondecode_error_is_retried_not_treated_as_local_validation
scripts/run_tests.sh tests/agent/test_error_classifier.py tests/tools/test_todo_tool.py tests/tools/test_registry.py tests/run_agent/test_run_agent.py::TestRetryExhaustion::test_jsondecode_error_is_retried_not_treated_as_local_validation

모두 통과했습니다.

비고

aideautomation/aide_hermes_agent 저장소는 코드 구조가 현재 hermes-agent와 상이하여 동일 커밋 체리픽이 불가했습니다.
실행 가능한 반영은 aideautomation/hermes-agent 포크 기준으로 완료했습니다.

Changed files

agent/error_classifier.py (modified, +24/-13)
run_agent.py (modified, +1/-1)
tests/agent/test_error_classifier.py (modified, +23/-0)
tests/run_agent/test_run_agent.py (modified, +20/-0)
tests/tools/test_registry.py (modified, +24/-0)
tests/tools/test_todo_tool.py (modified, +23/-0)
tools/registry.py (modified, +43/-1)
tools/todo_tool.py (modified, +16/-2)

PR #14349: fix(agent): inspect wrapped API error bodies

Repository: NousResearch/hermes-agent
Author: sgaofen
State: open | merged: False
Link: https://github.com/NousResearch/hermes-agent/pull/14349

Description (problem / solution / changelog)

Summary

Fixes #14195.
Makes _extract_error_body() walk wrapped exception cause/context chains, matching _extract_status_code().
Preserves top-level body precedence while allowing nested SDK bodies to drive retry/billing classification.

Root cause

Wrapped SDK exceptions could expose a nested HTTP status code while hiding the nested structured body. That made transient 402 usage-limit responses look like plain billing failures.

Tests

uv run --frozen --python 3.11 --extra dev pytest -o addopts='' tests/agent/test_error_classifier.py -q
git diff --check

Changed files

agent/error_classifier.py (modified, +19/-13)
tests/agent/test_error_classifier.py (modified, +38/-0)

RAW_BUFFERClick to expand / collapse

Summary

Affected files

agent/error_classifier.py:263-266
agent/error_classifier.py:774-788

Why this is a bug

At classification time:

status_code = _extract_status_code(error) walks the cause chain
body = _extract_error_body(error) does not walk the cause chain

So a wrapped error like this:

outer exception: Exception("outer")
__cause__: provider/SDK exception with status_code=402 and body message "Usage limit reached, try again in 5 minutes"

gets classified using:

status code: 402 (nested cause found)
message/body: outer / {} (nested body lost)

Minimal reproduction

Wrap a mock API error with:

status_code = 402
body = {"error": {"message": "Usage limit reached, try again in 5 minutes"}}

inside an outer exception and pass the outer exception to classify_error().

Expected

Classify as a transient/rate-limit condition, because the nested body explicitly says to retry later.

Actual

Classifies as billing, because the nested body is ignored and only the status code survives.

Suggested investigation

Make _extract_error_body() traverse __cause__ / __context__ the same way _extract_status_code() already does.

A regression test should cover wrapped 402/400 cases where only the nested body contains the decisive message.

extent analysis

TL;DR

Modify _extract_error_body() to walk the __cause__ and __context__ chain to extract the nested error body.

Guidance

Update _extract_error_body() to recursively check the __cause__ and __context__ attributes of the exception to find the nested error body.
Verify the fix by testing with a wrapped API error that has a nested body message, such as the minimal reproduction example provided.
Add a regression test to cover wrapped 402/400 cases where only the nested body contains the decisive message.
Review the classify_error() function to ensure it correctly handles the updated error body extraction.

Example

def _extract_error_body(error):
    if hasattr(error, '__cause__') and error.__cause__:
        return _extract_error_body(error.__cause__)
    elif hasattr(error, '__context__') and error.__context__:
        return _extract_error_body(error.__context__)
    # existing implementation to extract error body

Notes

This fix assumes that the nested error body is the first one found in the __cause__ or __context__ chain. If there are multiple nested error bodies, additional logic may be needed to determine which one to use.

Recommendation

Apply the workaround by modifying _extract_error_body() to walk the __cause__ and __context__ chain, as this will allow the function to correctly extract the nested error body and classify errors as transient or billing-related.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #device allocation #model download #tokenizer error #prompt formatting

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

hermes - ✅(Solved) Fix Wrapped API errors lose nested body details and misclassify transient 402s as billing failures [4 pull requests, 1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Expected

Fix Action

Fixed

PR fix notes

PR #14219: fix(agent): preserve nested error bodies during classification

Description (problem / solution / changelog)

Summary

Root cause

Fix

Regression coverage

Testing

Changed files

PR #14287: fix(agent): preserve nested API error bodies

Description (problem / solution / changelog)

Summary

Testing

Changed files

PR #2: fix: apply P1 issue hardening (todo/registry/retry)

Description (problem / solution / changelog)

결론

반영 이슈

변경 요약

테스트

비고

Changed files

PR #14349: fix(agent): inspect wrapped API error bodies

Description (problem / solution / changelog)

Summary

Root cause

Tests

Changed files

Summary

Affected files

Why this is a bug

Minimal reproduction

Expected

Actual

Suggested investigation

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

RELATED_DISCOVERY

TRENDING