mi/werkrevisor
1
0
Fork 0
Code Issues Pull Requests 2 Actions Packages Projects Releases Wiki Activity

werkrevisor 0.13.0: mechanical file changes skip the model #11

Merged
mi merged 9 commits from claude-code/mechanical-changes-skip-model into main 2026-10-05 19:28:32 +02:00
Conversation 2 Commits 9 Files Changed 23
+1038 -69
claude-code commented 2026-10-05 19:24:58 +02:00
Collaborator
Copy Link
Copy Source

werkrevisor 0.13.0: mechanical file changes skip the model.

  • A file is mechanical only if every changed line is a pure rename, a whitespace change (not in whitespace-significant files) or an import line of a known language; it goes neither into the diff nor into the full files for the model.
  • A PR of only mechanical files is GREEN-eligible without a model call; hard rules still apply.
  • Additional changes: the example host configs match mih09 (mode = "combined", token_count_factor = 1.55); an unknown change kind in a changed .werkrevisor.toml gives YELLOW, not RED.

Details: the PR document under docs/prs/.

🤖 Generated with Claude Code

werkrevisor 0.13.0: mechanical file changes skip the model. - A file is mechanical only if every changed line is a pure rename, a whitespace change (not in whitespace-significant files) or an import line of a known language; it goes neither into the diff nor into the full files for the model. - A PR of only mechanical files is GREEN-eligible without a model call; hard rules still apply. - Additional changes: the example host configs match mih09 (`mode = "combined"`, `token_count_factor = 1.55`); an unknown change kind in a changed `.werkrevisor.toml` gives YELLOW, not RED. Details: the PR document under `docs/prs/`. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
claude-code added 8 commits 2026-10-05 19:24:59 +02:00
docs: PR#11 document for mechanical changes that skip the model efbe99c9c7 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
mechanical file changes skip the model edd1dd6f76 
A changed file whose every changed line is an import, a whitespace change or a rename
is left out of the PR-Agent prompt and the full files, does not count for the coverage check,
and is listed in one line of the comment.
A PR of such files alone gets the change kind mechanical_only and can become GREEN without a model call.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
config: [mechanical] in every config, example host configs synced with mih09 248f3d6e74 
The example configs allow mechanical_only.
werkprobe, werkrevisor and hsadmin-ng use engine mode combined and token_count_factor 1.55,
measured on werkrevisor PR#10.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
mechanical files do not count for the size limits 564ee30276 
A package move across many files needs no AI judgement, so it does not make a PR too large.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
werkrevisor 0.13.0: README and docs describe [mechanical] f51e47f629 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
docs: PR#11 document completed with scenarios, solution and decisions 92014add0a 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
.werkrevisor.toml stays unchanged until 0.13.0 is deployed e4957f555e 
The installed 0.12.0 would reject mechanical_only; a follow-up PR adds it after the deploy.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
an unknown change kind in a changed .werkrevisor.toml gives YELLOW, not RED
ai-review/assessment AI findings; a developer approval after checking them unlocks the merge; review requested from mi
Details
werkator build succeeded after 00:23
Details
ai-review/gate build succeeded after 00:23
Details
ccad3614c1 
A newer werkrevisor may know it, as it may know an unknown key.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
claude-code marked the pull request as ready for review 2026-10-05 19:24:59 +02:00
ai-reviewer commented 2026-10-05 19:25:31 +02:00
Collaborator
Copy Link
Copy Source

🟠 werkrevisor: ORANGE / FINDINGS

Change kind: mixed.

Reasons

  • YELLOW test changes are not purely additive: tests/fakes.py:170 changes an existing test line, tests/test_classifier.py:67 changes an existing test line, tests/test_classifier.py:71 removes a test or assertion and 13 more
  • YELLOW change kind 'mixed' (main code, tests, other files, docs); AI-approvable are only: docs only, tests only added, docs and tests, refactoring with unchanged tests, feature with purely additive tests, harmless files only
  • YELLOW risky paths changed: config/hsadmin-ng.toml (config/), config/werkprobe.toml (config/), config/werkrevisor.toml (config/**) and 1 more
  • YELLOW spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 hard_to_read (0.60 > yellow_above = 0.5)
  • YELLOW spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 hard_to_read (0.54 > yellow_above = 0.5)
  • ORANGE spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 vague_then (0.92 > red_above = 0.85)
  • YELLOW 9 changed files were given to the AI only as diff, not in full, so it could not check them against the rest of their file (all full files together would exceed [review] max_full_files_chars = 120000): tests/test_rules.py, tests/test_token_count.py, werkrevisor/cli.py and 6 more
  • YELLOW no AI review: the PR needs 115477 input tokens (74501 counted with o200k_base × [model] token_count_factor = 1.55), more than the limit of 80000 derived from [budget] max_usd_per_review = 0.40

Classifier

Unit Question Problem Confidence Verdict
docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 hard_to_read 0.60 0.76 YELLOW
docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 hard_to_read 0.54 0.45 YELLOW
docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 vague_then 0.92 – ORANGE

18 of 20 units passed all questions (1 PR document(s)).
7 answers ignored because of a confidence below [classifier] min_confidence.

Usage: model `openrouter/anthropic/claude-sonnet-5.5`, LLM skipped (too large), tokens counted before the call 115477, duration 19.7 s, no model call, cost 0 USD, classifier `jev-1.13.0` calls 20, tokens in 18550, cost 0.0008 USD, config: defaults + host + repository (.werkrevisor.toml at base fcab521).
Reviewed head `ccad3614c19948967d1f07aa548a8bb5f7b45c10` against base `fcab52151f1c9b803c4d4ff8bffde3a924411cfd`, werkrevisor 0.12.0.
<!-- werkrevisor:ccad3614c19948967d1f07aa548a8bb5f7b45c10 --> <!-- werkrevisor-verdict:ORANGE --> ## 🟠 werkrevisor: ORANGE / FINDINGS Change kind: `mixed`. ### Reasons - **YELLOW** test changes are not purely additive: tests/fakes.py:170 changes an existing test line, tests/test_classifier.py:67 changes an existing test line, tests/test_classifier.py:71 removes a test or assertion and 13 more - **YELLOW** change kind 'mixed' (main code, tests, other files, docs); AI-approvable are only: docs only, tests only added, docs and tests, refactoring with unchanged tests, feature with purely additive tests, harmless files only - **YELLOW** risky paths changed: config/hsadmin-ng.toml (config/**), config/werkprobe.toml (config/**), config/werkrevisor.toml (config/**) and 1 more - **YELLOW** spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 hard_to_read (0.60 > yellow_above = 0.5) - **YELLOW** spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 hard_to_read (0.54 > yellow_above = 0.5) - **ORANGE** spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 vague_then (0.92 > red_above = 0.85) - **YELLOW** 9 changed files were given to the AI only as diff, not in full, so it could not check them against the rest of their file (all full files together would exceed [review] max_full_files_chars = 120000): tests/test_rules.py, tests/test_token_count.py, werkrevisor/cli.py and 6 more - **YELLOW** no AI review: the PR needs 115477 input tokens (74501 counted with o200k_base × [model] token_count_factor = 1.55), more than the limit of 80000 derived from [budget] max_usd_per_review = 0.40 ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 | hard_to_read | 0.60 | 0.76 | YELLOW | | docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 | hard_to_read | 0.54 | 0.45 | YELLOW | | docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.11 | vague_then | 0.92 | – | ORANGE | 18 of 20 units passed all questions (1 PR document(s)). 7 answers ignored because of a confidence below [classifier] min_confidence. <sub> Usage: model `openrouter/anthropic/claude-sonnet-5.5`, LLM skipped (too large), tokens counted before the call 115477, duration 19.7 s, no model call, cost 0 USD, classifier `jev-1.13.0` calls 20, tokens in 18550, cost 0.0008 USD, config: defaults + host + repository (.werkrevisor.toml at base fcab521).<br> Reviewed head `ccad3614c19948967d1f07aa548a8bb5f7b45c10` against base `fcab52151f1c9b803c4d4ff8bffde3a924411cfd`, werkrevisor 0.12.0. </sub>
ai-reviewer requested review from mi 2026-10-05 19:25:31 +02:00
claude-code added 1 commit 2026-10-05 19:26:34 +02:00
docs: Scenario#11.11 states the size limit and an observable Then, as Jev asked
ai-review/assessment human review required; a human approval unlocks the merge; a human review request is pending
Details
werkator build succeeded after 00:07
Details
ai-review/gate build succeeded after 00:08
Details
28567bcb9e 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ai-reviewer commented 2026-10-05 19:26:46 +02:00
Collaborator
Copy Link
Copy Source

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

  • YELLOW test changes are not purely additive: tests/fakes.py:170 changes an existing test line, tests/test_classifier.py:67 changes an existing test line, tests/test_classifier.py:71 removes a test or assertion and 13 more
  • YELLOW change kind 'mixed' (main code, tests, other files, docs); AI-approvable are only: docs only, tests only added, docs and tests, refactoring with unchanged tests, feature with purely additive tests, harmless files only
  • YELLOW risky paths changed: config/hsadmin-ng.toml (config/), config/werkprobe.toml (config/), config/werkrevisor.toml (config/**) and 1 more
  • YELLOW spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 hard_to_read (0.60 > yellow_above = 0.5)
  • YELLOW spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.13 hard_to_read (0.52 > yellow_above = 0.5)
  • YELLOW 9 changed files were given to the AI only as diff, not in full, so it could not check them against the rest of their file (all full files together would exceed [review] max_full_files_chars = 120000): tests/test_rules.py, tests/test_token_count.py, werkrevisor/cli.py and 6 more
  • YELLOW no AI review: the PR needs 115462 input tokens (74491 counted with o200k_base × [model] token_count_factor = 1.55), more than the limit of 80000 derived from [budget] max_usd_per_review = 0.40

Classifier

Unit Question Problem Confidence Verdict
docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 hard_to_read 0.60 0.75 YELLOW
docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.13 hard_to_read 0.52 0.52 YELLOW

18 of 20 units passed all questions (1 PR document(s)).
7 answers ignored because of a confidence below [classifier] min_confidence.

Usage: model `openrouter/anthropic/claude-sonnet-5.5`, LLM skipped (too large), tokens counted before the call 115462, duration 6.2 s, no model call, cost 0 USD, classifier `jev-1.13.0` calls 20, tokens in 18613, cost 0.0008 USD, config: defaults + host + repository (.werkrevisor.toml at base fcab521).
Reviewed head `28567bcb9eab45407f82913c1f061f3a6a88ebb5` against base `fcab52151f1c9b803c4d4ff8bffde3a924411cfd`, werkrevisor 0.12.0.
<!-- werkrevisor:28567bcb9eab45407f82913c1f061f3a6a88ebb5 --> <!-- werkrevisor-verdict:YELLOW --> ## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** test changes are not purely additive: tests/fakes.py:170 changes an existing test line, tests/test_classifier.py:67 changes an existing test line, tests/test_classifier.py:71 removes a test or assertion and 13 more - **YELLOW** change kind 'mixed' (main code, tests, other files, docs); AI-approvable are only: docs only, tests only added, docs and tests, refactoring with unchanged tests, feature with purely additive tests, harmless files only - **YELLOW** risky paths changed: config/hsadmin-ng.toml (config/**), config/werkprobe.toml (config/**), config/werkrevisor.toml (config/**) and 1 more - **YELLOW** spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 hard_to_read (0.60 > yellow_above = 0.5) - **YELLOW** spec: docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.13 hard_to_read (0.52 > yellow_above = 0.5) - **YELLOW** 9 changed files were given to the AI only as diff, not in full, so it could not check them against the rest of their file (all full files together would exceed [review] max_full_files_chars = 120000): tests/test_rules.py, tests/test_token_count.py, werkrevisor/cli.py and 6 more - **YELLOW** no AI review: the PR needs 115462 input tokens (74491 counted with o200k_base × [model] token_count_factor = 1.55), more than the limit of 80000 derived from [budget] max_usd_per_review = 0.40 ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.04 | hard_to_read | 0.60 | 0.75 | YELLOW | | docs/prs/2026-10-05-PR#11-mechanical-changes-skip-the-model.md Scenario#11.13 | hard_to_read | 0.52 | 0.52 | YELLOW | 18 of 20 units passed all questions (1 PR document(s)). 7 answers ignored because of a confidence below [classifier] min_confidence. <sub> Usage: model `openrouter/anthropic/claude-sonnet-5.5`, LLM skipped (too large), tokens counted before the call 115462, duration 6.2 s, no model call, cost 0 USD, classifier `jev-1.13.0` calls 20, tokens in 18613, cost 0.0008 USD, config: defaults + host + repository (.werkrevisor.toml at base fcab521).<br> Reviewed head `28567bcb9eab45407f82913c1f061f3a6a88ebb5` against base `fcab52151f1c9b803c4d4ff8bffde3a924411cfd`, werkrevisor 0.12.0. </sub>
mi merged commit d778c113e9 into main 2026-10-05 19:28:32 +02:00
mi deleted branch claude-code/mechanical-changes-skip-model 2026-10-05 19:28:32 +02:00
Sign in to join this conversation.
Reviewers
No Reviewers
mi
Labels
No items
No labels
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
ai-reviewer claude-code mi
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: mi/werkrevisor#11