fix(tests): keep __pycache__ out of a clean checkout

[thewrz]

This was written agentically; verify its assertions:

Why

A green tests/run-tests.sh run left agentkit/hooks/lib/__pycache__/ behind, so agent-run.sh --cmd test reported the checkout dirty and finding-ledger.sh evidence refused to record the run.

What

.gitignore now excludes __pycache__/ and *.pyc. tests/run-tests.sh exports PYTHONDONTWRITEBYTECODE=1 as defense in depth. The actual write path was test-rewrite-runtime.sh's --prefix-fixture reinvocation, which launches python3 -I (implies -E, so it ignores PYTHONDONTWRITEBYTECODE) without -B; it now passes -B too. A new suite, tests/test-clean-tree-after-probe.sh, pins the fix: it fails on the unfixed fixture and passes after.

Testing

• tests/run-tests.sh green
• new test fails on origin/main, passes here
• agent-run.sh --cmd test exits 0 with a clean tree, PYTHONDONTWRITEBYTECODE unset
• CI green

🤖 Co-authored by Claude Fable 5.1 (plan) and Claude Sonnet 5 (implementation). Closes #897.

🤖 Generated with Claude Code

Summary by CodeRabbit

• Chores
  • Python test runs now avoid creating bytecode cache files, helping keep the project directory clean.
  • Added a check to verify that running a runtime probe does not leave Python cache directories behind.

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Team

Run ID: 4112a646-c57a-47e6-bbd4-1fab9828637c

📥 Commits

Reviewing files that changed from the base of the PR and between d4ba2f8 and 3a3949b.

📒 Files selected for processing (4)

• .gitignore
• tests/fixtures/test_rewrite_runtime.py
• tests/run-tests.sh
• tests/test-clean-tree-after-probe.sh

Included review availability: 9 reviews are currently available. Your included PR review attempts over the past 7 days set your current allowance at 10 reviews per hour.

---

📝 Walkthrough

Walkthrough

The test runner and rewrite-runtime fixture now disable Python bytecode writes. A new shell test checks for __pycache__ directories before and after the fixture runs. .gitignore excludes Python cache directories and compiled files.

Changes

Python bytecode cleanliness

Layer / File(s)	Summary
Disable and check Python cache writes tests/run-tests.sh, tests/fixtures/test_rewrite_runtime.py, tests/test-clean-tree-after-probe.sh, .gitignore	The test runner sets PYTHONDONTWRITEBYTECODE=1, and the fixture uses Python’s -B flag. A new test checks for __pycache__ directories before and after fixture execution and verifies its exit status. .gitignore excludes __pycache__/ and *.pyc.

Priority: ➖ Normal

Change: Bug fix · Severity of issue fixed: Medium

Merge Risk: ⚪ Minimal · up to 3a394

No actionable merge risk remains; the new cleanliness test is included in the normal test run.

🚥 Pre-merge checks | ✅ 3 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Linked Issues check	⚠️ Warning	PR #899 implements the main fixes for #897: .gitignore ignores __pycache__/ and *.pyc, tests/run-tests.sh exports PYTHONDONTWRITEBYTECODE=1, and the isolated fixture uses python3 -B. The a…	Update tests/test-clean-tree-after-probe.sh to run the relevant probe coverage and assert an empty git status --porcelain after the run. Keep the focused __pycache__ assertion if it remains useful.
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 3 files. (1 skipped: 1…	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Out of Scope Changes check	✅ Passed	The reviewed changes are limited to Python bytecode exclusion, test-process bytecode prevention, the isolated fixture invocation, and regression coverage for issue #897. No unrelated change is establi…
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely describes the primary change: preventing pycache files from leaving the test checkout dirty.

Full details: Linked Issues check

Explanation

PR #899 implements the main fixes for #897: .gitignore ignores __pycache__/ and *.pyc, tests/run-tests.sh exports PYTHONDONTWRITEBYTECODE=1, and the isolated fixture uses python3 -B. The added test runs the fixture and checks for __pycache__ under agentkit. It does not assert that git status --porcelain is empty after the probe, as required by #897, and it does not cover other checkout changes.

Full details: Docstring Coverage

Explanation

Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 3 files. (1 skipped: 1 unsupported.)

---

_{Comment @coderabbitai help to get the list of available commands.}

[thewrz]

This was written agentically; verify its assertions:

Adversarial review receipt

• Reviewer: provider=openai; model=gpt-6-astra; effort=xhigh; mode=cross-provider (reason: n/a)
• Launcher: adversarial-run.sh sha256=4d1fc623db1b387d5d6dc0792b56ea5fcc25bf1297a77b9aca4dab243902bd50; attempt=4ae5f6c6-c113-4200-a923-1f5c77c98d1e
• Procedure: one-shot diff review; no contract-blind or two-pass attestation
• Reviewer override: configured=gpt-6-astra-xhigh; operator-authorized=gpt-6-astra-xhigh
• Counts: P1=0; P2=0; total=0
• Reviewed head: 3a3949b
• Diff payload: wrzonance/agent-kit:899:5fe49204b3157c9d4fac18a74f914c8f9c304286fe552acd08fcf19e373f36bb
• Confirmed finding: none confirmed
• Execution: performed; adjudication=recorded
• Remediation: complete

🤖 Co-authored by Claude Fable 5.1 noreply@anthropic.com.

[thewrz]

This was written agentically; verify its assertions:

Review ledger

Machine-readable record of every review already performed on this PR.

{
  "version": 1,
  "pr": 899,
  "repo": "wrzonance/agent-kit",
  "reviews": [
    {
      "kind": "adversarial",
      "provider": "openai",
      "model": "gpt-6-astra",
      "effort": "xhigh",
      "mode": "cross-provider",
      "attemptId": "4ae5f6c6-c113-4200-a923-1f5c77c98d1e",
      "launcherSha256": "4d1fc623db1b387d5d6dc0792b56ea5fcc25bf1297a77b9aca4dab243902bd50",
      "procedure": "one-shot diff review; no contract-blind or two-pass attestation",
      "reviewerOverride": "configured=gpt-6-astra-xhigh; operator-authorized=gpt-6-astra-xhigh",
      "harness": "codex",
      "head_sha": "3a3949b80b17ff03815f0e39cb90551b17c61d75",
      "covered_heads": [
        "3a3949b80b17ff03815f0e39cb90551b17c61d75"
      ],
      "diff_payload": "wrzonance/agent-kit:899:5fe49204b3157c9d4fac18a74f914c8f9c304286fe552acd08fcf19e373f36bb",
      "findings": [],
      "counts": {
        "p1": 0,
        "p2": 0
      },
      "reviewed_at": "2026-09-24T04:57:28Z"
    }
  ]
}

🤖 Co-authored by Claude Fable 5.1 noreply@anthropic.com.

[thewrz]

@coderabbitai full review

[coderabbitai]

✅ Action performed

Full review finished.