csv_output silently overwrites distinct fields when flattened keys collide

Open Beginner friendly
#3,865 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 2 days

Nobody has claimed this yet.

Assessment

Difficulty
2/5
Estimated time
1-3 hours
Newbie friendliness
84/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Active
Tech stack
go
Domain
backend

Research direction

Read pkg/github/csv_output.go, especially appendFlattenedCSVFields and csvColumnName, then review the flattening tests in pkg/github/csv_output_test.go. Add the reported JSON row with a literal dotted key and its equivalent nested path. Done means the output is deterministic and preserves both values, or clearly reports that the row cannot be represented without loss.

Written by the indexing model from the issue text.

Description

Summary

When csv_output is enabled, list-tool results are converted to CSV by flattening nested objects with .-joined column names. A literal field name containing . can produce the same column name as a nested field. One value is silently overwritten; Go map iteration makes the surviving value unpredictable.

Reproduction (from the current implementation)

Convert a list response containing this JSON row through jsonTextToCSV:

[{"user.login":"literal","user":{"login":"nested"}}]

appendFlattenedCSVFields maps both the top-level user.login and the nested user → login to the same row["user.login"]. The resulting CSV has only one user.login column, containing either literal or nested, depending on map iteration order. The other value is lost.

The conversion is used by withCSVOutput for default list_* tools whenever the csv_output feature flag is enabled (including insiders mode). This is a code-path demonstration, not a claim of a live GitHub API response with these exact fields.

Expected behavior

Distinct JSON fields remain distinguishable in CSV, or the conversion explicitly reports that it cannot represent this row without loss. The chosen output should be deterministic.

Suggested fix

Escape/encode path components when building flattened column names, or detect collisions and use unambiguous names/fail clearly. Add a regression test with both a literal dotted key and an equivalent nested path (and verify neither value is silently dropped).

Source
  • pkg/github/csv_output.go: appendFlattenedCSVFields, csvColumnName, csvRowsFromArray, and withCSVOutput.
  • pkg/github/csv_output_test.go: existing flattening tests cover nested objects, but not colliding flattened names.
Dominant language
Go
Stars
34.4k
Forks
5.3k
Avg merge
3d 10h
Merged PRs (30d)
11

Getting set up

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 github/github-mcp-server

All issues in github/github-mcp-server

Similar issues

More Go issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.