werkrevisor 0.14.0: severity of the model findings #14

2026-10-05T20:21:29+02:00

claude-code commented

2026-10-05 20:21:29 +02:00

werkrevisor 0.14.0: the severity of the model findings decides between ORANGE and YELLOW.

The model marks every finding as [high], [medium] or [low]; werkrevisor reads and strips the mark.
ORANGE needs one high finding or [review] orange_min_medium (default 2) medium ones; other findings give YELLOW.
[review] orange_min_medium replaces red_min_findings.

Details: the PR document under docs/prs/.

🤖 Generated with Claude Code

werkrevisor 0.14.0: the severity of the model findings decides between ORANGE and YELLOW. - The model marks every finding as `[high]`, `[medium]` or `[low]`; werkrevisor reads and strips the mark. - ORANGE needs one high finding or `[review] orange_min_medium` (default 2) medium ones; other findings give YELLOW. - `[review] orange_min_medium` replaces `red_min_findings`. Details: the PR document under `docs/prs/`. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-code added 4 commits 2026-10-05 20:21:29 +02:00

docs: PR#14 document for the severity of the model findings cd35d3562b

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

werkrevisor 0.14.0: the severity of the model findings decides between ORANGE and YELLOW 174969afc1

The operator instructions ask the model to start every issue_header with
[high], [medium] or [low]; werkrevisor parses and strips the prefix.
One high finding or [review] orange_min_medium medium findings give ORANGE,
other findings give YELLOW; a missing or unknown severity counts as medium.
The comment lists the findings from high to low severity.
[review] orange_min_medium replaces [review] red_min_findings.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

docs: README and operations doc describe the severity of the model findings 1cdb64b8b5

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

docs: PR#14 document names the deploy order and the adapted test fixtures

werkator build succeeded after 00:18

Details

0eeb8e5524

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

claude-code marked the pull request as ready for review 2026-10-05 20:36:56 +02:00

claude-code added 1 commit 2026-10-05 20:36:56 +02:00

Merge remote-tracking branch 'origin/main' into claude-code/finding-severity

ai-review/assessment human review required; a human approval unlocks the merge; review requested from mi

Details

werkator build succeeded after 00:08

Details

ai-review/gate build succeeded after 00:23

Details

a6ff8c795a

# Conflicts:
#	README.md
#	docs/betrieb-git.javagil.de.md
#	werkrevisor/__init__.py

ai-reviewer commented

2026-10-05 20:37:22 +02:00

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

YELLOW test changes are not purely additive: tests/fakes.py:141 changes an existing test line, tests/fakes.py:147 changes an existing test line, tests/test_classifier.py:318 changes an existing test line and 18 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, mechanical changes only
YELLOW risky paths changed: .werkrevisor.toml (built in), config/hsadmin-ng.toml (config/), config/werkprobe.toml (config/) and 1 more
YELLOW the changed .werkrevisor.toml uses settings unknown to the installed werkrevisor 0.13.2: [review] orange_min_medium; deploy a version that knows them right after the merge, or every review will end with ERROR
YELLOW 5 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_flow.py, werkrevisor/cli.py, werkrevisor/config.py and 2 more
YELLOW no AI review: the PR needs 82285 input tokens (53087 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

13 of 13 units passed all questions (1 PR document(s)).
2 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 82285, duration 17.9 s, no model call, cost 0 USD, classifier `jev-1.13.0` calls 13, tokens in 11701, cost 0.0005 USD, config: defaults + host + repository (.werkrevisor.toml at base ecad961).

Reviewed head `a6ff8c795a92dfccd6030c35ee8bee1767dfdfd7` against base `ecad961fe09b87c20ae30836a09e1de8cab23502`, werkrevisor 0.13.2.}

## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** test changes are not purely additive: tests/fakes.py:141 changes an existing test line, tests/fakes.py:147 changes an existing test line, tests/test_classifier.py:318 changes an existing test line and 18 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, mechanical changes only - **YELLOW** risky paths changed: .werkrevisor.toml (built in), config/hsadmin-ng.toml (config/**), config/werkprobe.toml (config/**) and 1 more - **YELLOW** the changed .werkrevisor.toml uses settings unknown to the installed werkrevisor 0.13.2: [review] orange_min_medium; deploy a version that knows them right after the merge, or every review will end with ERROR - **YELLOW** 5 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_flow.py, werkrevisor/cli.py, werkrevisor/config.py and 2 more - **YELLOW** no AI review: the PR needs 82285 input tokens (53087 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 13 of 13 units passed all questions (1 PR document(s)). 2 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 82285, duration 17.9 s, no model call, cost 0 USD, classifier `jev-1.13.0` calls 13, tokens in 11701, cost 0.0005 USD, config: defaults + host + repository (.werkrevisor.toml at base ecad961).<br> Reviewed head `a6ff8c795a92dfccd6030c35ee8bee1767dfdfd7` against base `ecad961fe09b87c20ae30836a09e1de8cab23502`, werkrevisor 0.13.2. </sub>

ai-reviewer requested review from mi 2026-10-05 20:37:23 +02:00

mi merged commit e8b256ef92 into main

2026-10-05 20:39:30 +02:00

mi deleted branch claude-code/finding-severity

2026-10-05 20:39:30 +02:00

Sign in to join this conversation.

2 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: mi/werkrevisor#14