feat(codex): Codex harness backend - #11
Merged
Merged
Conversation
RobertCrupa
added this pull request to stack #12
September 21, 2026 13:26
RobertCrupa
force-pushed
the
feat/codex-harness
branch
from
September 21, 2026 13:31
8c3df27 to
cea40b4
Compare
artogahr
approved these changes
Sep 21, 2026
RobertCrupa
force-pushed
the
feat/codex-harness
branch
from
September 27, 2026 08:09
4d6ec77 to
b8acff3
Compare
RobertCrupa
force-pushed
the
feat/codex-harness
branch
3 times, most recently
from
September 29, 2026 12:45
d57c6c0 to
eb9d3f6
Compare
Adds Codex as a second harness adapter alongside Claude Code: isolated sessions with MCP and process limits (temp CODEX_HOME, empty working directory, shell tools disabled, read-only sandbox, pre-approved allowlisted MCP tools so call-actor is not rejected), a JSONL event parser with captured fixtures, and workspace test/documentation wiring. Builds on feat/scenario-suites.
The repo's own suites write `input.prompt`; the mcp-server-evals datasets
write `input.query`. Reading only `prompt` fell through to
`JSON.stringify(input)`, so an item from those datasets handed the agent
the literal string `{"query":"..."}` and still produced a trace, a rubric
verdict and scores that all looked valid.
Temporary. The fix is one key across every dataset, asserted at sync time
rather than reinterpreted at run time: apify/ai-team#327. Drop the `query`
fallback once the datasets are normalised.
RobertCrupa
force-pushed
the
feat/codex-harness
branch
from
September 29, 2026 12:46
eb9d3f6 to
9f686c2
Compare
apify-service-account
added
the
tested
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds Codex as a second harness adapter alongside Claude Code: isolated sessions with MCP and process limits (temp `CODEX_HOME`, empty working directory, shell tools disabled, read-only sandbox, pre-approved allowlisted MCP tools so `call-actor` isn't rejected), a JSONL event parser with captured fixtures, and the runner/docs wiring to select it via `harness.kind`.
Depends on scenario suites from #10. Last PR in this stack.