Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

chore: remove silent exception swallowing (fail-loudly debt)

Open
#4,239 1 comment 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
35/100
Issue type
Refactor
Clarity
Mostly clear
Activity status
Active
Tech stack
python
Domain
ai, frontend, tooling

Research direction

Start with CLAUDE.md and the listed handlers in agents/base/memory.py, agents/base/agent.py, agents/tools/file_io_tools.py, ui/_chat_helpers.py, and ui/routers/system.py; review related issue #3289. Trace which failures are silently swallowed, then verify behavior-relevant sites no longer hide errors and a lint test rejects new silent broad handlers while allowing documented existing locations.

Written by the indexing model from the issue text.

Description

enhancement p2

Several hundred broad except Exception handlers remain, and a meaningful share silently discard the error — the pattern CLAUDE.md's fail-loudly rule calls out. Some hide real behaviour: memory ranking silently stops adapting, the "model loaded at wrong ctx" check silently turns off, and every Agent UI streaming failure becomes a bare "try again" with no cause or correlation id.

Next step: fix the behaviour-relevant sites first, then add a lint guard with an allowlist of what's left.

🔍 Technical details

Highest-value sites:

  • agents/base/memory.py:1466 (failed update_confidence swallowed), :2022 (vectors dropped)
  • agents/base/agent.py:4927 (ctx-size probe error → False)
  • agents/tools/file_io_tools.py:725 (search_code skips files)
  • ui/_chat_helpers.py:2787-2796 (streaming failure → generic message; failed persist swallowed); also :93, :1672, :2250, :2821
  • ui/routers/system.py:606 (catalog failure leaves model_downloaded wrong)

Counts: agents/base 47 swallowing sites (19 in system_context.py); src/gaia/ui 26 of 141 broad handlers only pass/continue/return a constant; llm+installer 108 broad (21 swallow); rag 40, eval 37, audio 33, cli 64, hub 16.

Guard: a lint test failing on new silent broad handlers, allowlisting existing ones by location. Related: #3289.

Found in a review of main @ 64544f218.

Dominant language
Python
Stars
1.6k
Forks
169
Avg merge
2d 21h
Merged PRs (30d)
489

Getting set up

Open in Codespaces

Starts the project's dev container in your browser, under your own GitHub account.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from amd/gaia

All issues in amd/gaia

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.