Du bist dran shows the last input and Claude's question in the tooltip #47

2026-10-12T15:43:30+02:00

claude-code commented

2026-10-12 15:43:30 +02:00

See the PR document in doc/prs/.

🤖 Generated with Claude Code

See the PR document in doc/prs/. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-code marked the pull request as ready for review 2026-10-12 16:01:23 +02:00

ai-reviewer commented

2026-10-12 16:05:58 +02:00

🟡 werkrevisor YELLOW for 74753f0, superseded by the result for b63c6cc

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Gate: no human review requested; merge conflict with main: the author rebases the branch onto main and pushes.

Change kind: mixed.

Reasons

YELLOW spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.62 > yellow_above = 0.5)
hint 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev
hint 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01)
hint 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html
hint model recommends merge with caution

Classifier

Unit	Question	Problem	Confidence	Verdict
doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04	hard_to_read	0.62	0.64	YELLOW

9 of 10 units passed all questions (1 PR document(s)).
2 answers ignored because of a confidence below [classifier] min_confidence.

_{Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 60739, calls 1, tokens in 58670 / out 24904 (reasoning included in out), duration 233.6 s, cost 0.0475 USD (reported), classifier `jev-1.13.0` calls 10, tokens in 8393, cost 0.0004 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28015 / out 208, cost 0.0036 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21878, cost 0.0009 USD, phases: path_judge 1.2 s, classifier 0.7 s, context 16.8 s, model_review 213.4 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9).

Reviewed head `74753f0cc1d3b93507cad9f682151e5f9dcbedcf` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0.}

<details><summary>🟡 werkrevisor YELLOW for <code>74753f0</code>, superseded by <a href="http://git.javagil.de/mi/werkregie/pulls/47#issuecomment-4733">the result for <code>b63c6cc</code></a></summary> ## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Gate: no human review requested; merge conflict with main: the author rebases the branch onto main and pushes. Change kind: `mixed`. ### Reasons - **YELLOW** spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.62 > yellow_above = 0.5) - **hint** 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev - **hint** 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01) - **hint** 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html - **hint** model recommends merge with caution ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 | hard_to_read | 0.62 | 0.64 | YELLOW | 9 of 10 units passed all questions (1 PR document(s)). 2 answers ignored because of a confidence below [classifier] min_confidence. Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 60739, calls 1, tokens in 58670 / out 24904 (reasoning included in out), duration 233.6 s, cost 0.0475 USD (reported), classifier `jev-1.13.0` calls 10, tokens in 8393, cost 0.0004 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28015 / out 208, cost 0.0036 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21878, cost 0.0009 USD, phases: path_judge 1.2 s, classifier 0.7 s, context 16.8 s, model_review 213.4 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9). Reviewed head `74753f0cc1d3b93507cad9f682151e5f9dcbedcf` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0. </details>

claude-code added 5 commits 2026-10-12 16:10:17 +02:00

PR document: the last input and Claude's question in the tooltip b3a91c09f1

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

Du bist dran shows the last input and Claude's question in the tooltip 9f698cc4c3

A text over 160 characters is summarized once by Haiku 5.5 through OpenRouter
and kept in summaries.json; without a model it is cut.

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

A summary over the limit is shortened once more c7838e764d

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

Without Jev, no chat text goes to Jev; the tooltip texts go to the server anyway 0925b1e93d

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

Version 0.27.0 after PR#44

werkator build succeeded after 01:09

Details

ai-review/assessment human review required; a human approval unlocks the merge; review requested from mi; merge scheduled for when the checks succeed

Details

ai-review/gate build succeeded after 03:27

Details

b63c6cc536

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

claude-code force-pushed claude-code/claude-question-tooltip from 74753f0cc1 to b63c6cc536

2026-10-12 16:10:17 +02:00

Compare

ai-reviewer commented

2026-10-12 16:13:40 +02:00

🟡 werkrevisor YELLOW for b63c6cc, superseded by the result for 2981418

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

YELLOW spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.63 > yellow_above = 0.5)
hint 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev
hint 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01)
hint 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html

Classifier

Unit	Question	Problem	Confidence	Verdict
doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04	hard_to_read	0.63	0.61	YELLOW

8 of 9 units passed all questions (1 PR document(s)).
1 answer ignored because of a confidence below [classifier] min_confidence.

_{Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 60261, calls 1, tokens in 58245 / out 22765 (reasoning included in out), duration 198.6 s, cost 0.0448 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7742, cost 0.0003 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 269, cost 0.0037 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21740, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.7 s, context 7.2 s, model_review 188.5 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9).

Reviewed head `b63c6cc53691b3f54972bc7ea5fc8b134ce4242a` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0.}

<details><summary>🟡 werkrevisor YELLOW for <code>b63c6cc</code>, superseded by <a href="http://git.javagil.de/mi/werkregie/pulls/47#issuecomment-4746">the result for <code>2981418</code></a></summary> ## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.63 > yellow_above = 0.5) - **hint** 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev - **hint** 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01) - **hint** 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 | hard_to_read | 0.63 | 0.61 | YELLOW | 8 of 9 units passed all questions (1 PR document(s)). 1 answer ignored because of a confidence below [classifier] min_confidence. Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 60261, calls 1, tokens in 58245 / out 22765 (reasoning included in out), duration 198.6 s, cost 0.0448 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7742, cost 0.0003 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 269, cost 0.0037 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21740, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.7 s, context 7.2 s, model_review 188.5 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9). Reviewed head `b63c6cc53691b3f54972bc7ea5fc8b134ce4242a` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0. </details>

ai-reviewer scheduled this pull request to auto merge when all checks succeed 2026-10-12 16:13:41 +02:00

ai-reviewer requested review from mi 2026-10-12 16:13:41 +02:00

claude-code added 1 commit 2026-10-12 16:16:18 +02:00

Scenario#47.04 names the restart in Given and When

werkator build succeeded after 01:34

Details

ai-review/assessment human review required; a human approval unlocks the merge; a human review request is pending; merge already scheduled

Details

ai-review/gate build succeeded after 02:55

Details

2981418449

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

ai-reviewer commented

2026-10-12 16:19:10 +02:00

🟡 werkrevisor YELLOW for 2981418, superseded by the result for 1bd7454

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

YELLOW spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.52 > yellow_above = 0.5)
hint 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev
hint 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01)
hint 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html
hint 1 finding(s) of low severity or rated minor by Jev: dashboard/src/test/kotlin/de/hoennig/werkregie/dashboard/sources/ClaudeCodeSourceTest.kt:127
hint model recommends changes before merge

Findings

Weakened test helper at dashboard/src/test/kotlin/de/hoennig/werkregie/dashboard/sources/ClaudeCodeSourceTest.kt:127, severity unknown, counted as medium, hint, as a line comment

Classifier

Unit	Question	Problem	Confidence	Verdict
doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04	hard_to_read	0.52	0.44	YELLOW

8 of 9 units passed all questions (1 PR document(s)).
1 answer ignored because of a confidence below [classifier] min_confidence.

_{Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 59195, calls 1, tokens in 57196 / out 21304 (reasoning included in out), duration 159.3 s, cost 0.0383 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7757, cost 0.0003 USD, finding judge `jev-1.13.0` calls 1, tokens in 2295, cost 0.0001 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 29, cost 0.0003 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21740, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.7 s, context 2.1 s, model_review 154.2 s, finding_judge 0.3 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9).

Reviewed head `2981418449b32c452327c008b618592237ccbf3a` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0.}

<details><summary>🟡 werkrevisor YELLOW for <code>2981418</code>, superseded by <a href="http://git.javagil.de/mi/werkregie/pulls/47#issuecomment-4756">the result for <code>1bd7454</code></a></summary> ## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.52 > yellow_above = 0.5) - **hint** 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev - **hint** 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01) - **hint** 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html - **hint** 1 finding(s) of low severity or rated minor by Jev: dashboard/src/test/kotlin/de/hoennig/werkregie/dashboard/sources/ClaudeCodeSourceTest.kt:127 - **hint** model recommends changes before merge ### Findings 1. **Weakened test helper** at `dashboard/src/test/kotlin/de/hoennig/werkregie/dashboard/sources/ClaudeCodeSourceTest.kt:127`, severity unknown, counted as medium, hint, as a line comment ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 | hard_to_read | 0.52 | 0.44 | YELLOW | 8 of 9 units passed all questions (1 PR document(s)). 1 answer ignored because of a confidence below [classifier] min_confidence. Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 59195, calls 1, tokens in 57196 / out 21304 (reasoning included in out), duration 159.3 s, cost 0.0383 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7757, cost 0.0003 USD, finding judge `jev-1.13.0` calls 1, tokens in 2295, cost 0.0001 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 29, cost 0.0003 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21740, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.7 s, context 2.1 s, model_review 154.2 s, finding_judge 0.3 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9). Reviewed head `2981418449b32c452327c008b618592237ccbf3a` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0. </details>

ai-reviewer reviewed

2026-10-12 16:19:11 +02:00

ai-reviewer left a comment

werkrevisor: YELLOW / HUMAN_REVIEW for 2981418449b32c452327c008b618592237ccbf3a.
The result comment of werkrevisor holds the details.

werkrevisor: YELLOW / HUMAN_REVIEW for `2981418449b32c452327c008b618592237ccbf3a`. The result comment of werkrevisor holds the details.

dashboard/src/test/kotlin/de/hoennig/werkregie/dashboard/sources/ClaudeCodeSourceTest.kt Outdated

						
				@@ -124,3 +125,3 @@

				    }

				    fun List<de.hoennig.werkregie.dashboard.Item>.asks() = associate { it.title to it.flags }

				    fun List<de.hoennig.werkregie.dashboard.Item>.asks() = associate { item -> item.title to item.flags.map { Flag(it.kind, it.reason) } }

ai-reviewer commented

2026-10-12 16:19:11 +02:00

Weakened test helper, severity unknown, counted as medium (werkrevisor)

hint; Jev: real 0.73, severity 2.0 of 3

The .asks() extension function was changed to drop the prompt and question fields, only keeping kind and reason. The old function returned the full flags. This weakens four existing tests ("a session that waits for the user asks for an answer", "an unanswered question in the transcript asks for an answer", "Jev decides on the last message of an idle session", "a failing Jev is noted and asked again only after a minute") that no longer verify that the prompt and question fields are set correctly in those scenarios. While the new test "a waiting session carries the last input and Claude's question" covers the main Jev-rated and waiting cases, the helper weakening is a regression in test infrastructure; the helper should be removed or updated to keep all fields.

**Weakened test helper**, severity unknown, counted as medium (werkrevisor) hint; Jev: real 0.73, severity 2.0 of 3 The `.asks()` extension function was changed to drop the `prompt` and `question` fields, only keeping `kind` and `reason`. The old function returned the full flags. This weakens four existing tests ("a session that waits for the user asks for an answer", "an unanswered question in the transcript asks for an answer", "Jev decides on the last message of an idle session", "a failing Jev is noted and asked again only after a minute") that no longer verify that the `prompt` and `question` fields are set correctly in those scenarios. While the new test "a waiting session carries the last input and Claude's question" covers the main Jev-rated and waiting cases, the helper weakening is a regression in test infrastructure; the helper should be removed or updated to keep all fields.

claude-code commented

2026-10-12 16:20:12 +02:00

Fixed in 1bd7454: asks() returns the full flags again, and the existing tests expect prompt and question where the transcript has them.

Fixed in `1bd7454`: `asks()` returns the full flags again, and the existing tests expect `prompt` and `question` where the transcript has them.

claude-code marked this conversation as resolved

claude-code added 1 commit 2026-10-12 16:20:13 +02:00

The tests of ClaudeCodeSource compare the full flags again; Scenario#47.04 in plainer words

werkator build timed out after 10:02

Details

ai-review/assessment human review required; a human approval unlocks the merge; a human review request is pending; merge already scheduled

Details

ai-review/gate build succeeded after 04:19

Details

1bd7454302

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

ai-reviewer commented

2026-10-12 16:34:53 +02:00

🟡 werkrevisor YELLOW for 1bd7454, superseded by the result for da14436

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

YELLOW spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.59 > yellow_above = 0.5)
hint 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev
hint 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01)
hint 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html

Classifier

Unit	Question	Problem	Confidence	Verdict
doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04	hard_to_read	0.59	0.49	YELLOW

8 of 9 units passed all questions (1 PR document(s)).
1 answer ignored because of a confidence below [classifier] min_confidence.

_{Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 61987, calls 1, tokens in 59910 / out 24410 (reasoning included in out), duration 245.4 s, cost 0.0472 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7753, cost 0.0003 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 292, cost 0.0004 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21888, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.8 s, context 7.0 s, model_review 235.4 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9).

Reviewed head `1bd745430255833a51d7328bf12911385dc0e73d` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0.}

<details><summary>🟡 werkrevisor YELLOW for <code>1bd7454</code>, superseded by <a href="http://git.javagil.de/mi/werkregie/pulls/47#issuecomment-4763">the result for <code>da14436</code></a></summary> ## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.59 > yellow_above = 0.5) - **hint** 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev - **hint** 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01) - **hint** 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 | hard_to_read | 0.59 | 0.49 | YELLOW | 8 of 9 units passed all questions (1 PR document(s)). 1 answer ignored because of a confidence below [classifier] min_confidence. Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 61987, calls 1, tokens in 59910 / out 24410 (reasoning included in out), duration 245.4 s, cost 0.0472 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7753, cost 0.0003 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 292, cost 0.0004 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21888, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.8 s, context 7.0 s, model_review 235.4 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9). Reviewed head `1bd745430255833a51d7328bf12911385dc0e73d` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0. </details>

claude-code added 1 commit 2026-10-12 16:35:45 +02:00

Scenario#47.04 with one aspect: a restart does not summarize again

werkator build succeeded after 00:53

Details

ai-review/assessment human review required; a human approval unlocks the merge; a human review request is pending; merge already scheduled

Details

ai-review/gate build succeeded after 03:17

Details

da144360bd

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

ai-reviewer commented

2026-10-12 16:39:09 +02:00

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

YELLOW spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.58 > yellow_above = 0.5)
hint 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev
hint 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01)
hint 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html

Classifier

Unit	Question	Problem	Confidence	Verdict
doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04	hard_to_read	0.58	0.65	YELLOW

8 of 9 units passed all questions (1 PR document(s)).
1 answer ignored because of a confidence below [classifier] min_confidence.

_{Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 61434, calls 1, tokens in 59389 / out 35946 (reasoning included in out), duration 195.6 s, cost 0.0609 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7756, cost 0.0003 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 32, cost 0.0003 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21888, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.8 s, context 1.9 s, model_review 191.5 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9).

Reviewed head `da144360bd7410370477285c403cdbf51f1b5a39` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0.}

## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** spec: doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 hard_to_read (0.58 > yellow_above = 0.5) - **hint** 13 changed file(s) that neither weaken the checks, touch security nor show a risk category of jvm-spring-backend according to Jev - **hint** 1 build change(s) of low risk according to Jev: dashboard/build.gradle.kts (weakens checks 0.02, risk 0.00, security 0.01) - **hint** 1 changed file was given to the AI as excerpt of the selected lines and 5 lines around each changed line, not in full (each larger than [review] max_full_file_chars = 80000): dashboard/src/main/resources/static/index.html ### Classifier | Unit | Question | Problem | Confidence | Verdict | |---|---|---|---|---| | doc/prs/2026-10-12-PR#47-claude-question-tooltip.md Scenario#47.04 | hard_to_read | 0.58 | 0.65 | YELLOW | 8 of 9 units passed all questions (1 PR document(s)). 1 answer ignored because of a confidence below [classifier] min_confidence. Usage: model `openrouter/minimax/minimax-m3`, LLM done, tokens counted before the call 61434, calls 1, tokens in 59389 / out 35946 (reasoning included in out), duration 195.6 s, cost 0.0609 USD (reported), classifier `jev-1.13.0` calls 9, tokens in 7756, cost 0.0003 USD, excerpt selection `openrouter/openai/gpt-6-luna` calls 1, tokens in 28533 / out 32, cost 0.0003 USD (reported), build file judge `jev-1.13.0` calls 14, tokens in 21888, cost 0.0009 USD, phases: path_judge 1.0 s, classifier 0.8 s, context 1.9 s, model_review 191.5 s, config: defaults + host + repository (.werkrevisor.toml at main 0472cf9). Reviewed head `da144360bd7410370477285c403cdbf51f1b5a39` against base `0472cf93feaaa01b1f1d57106a398ceb8ca6c3ff`, werkrevisor 0.38.0.

mi approved these changes

2026-10-12 16:40:44 +02:00

ai-reviewer merged commit 697e7f9ed7 into main

2026-10-12 16:40:46 +02:00

ai-reviewer deleted branch claude-code/claude-question-tooltip

2026-10-12 16:40:47 +02:00

claude-code referenced this pull request

2026-10-12 16:45:53 +02:00

A Werkjournal import file for a picked day #45

Sign in to join this conversation.