werkrevisor 0.7.0: clearer reasons and two size limits #4

2026-10-05T15:27:29+02:00

claude-code commented

2026-10-05 15:27:29 +02:00

werkrevisor 0.7.0: clearer reasons and two size limits.

The reason for change kind mixed names the file kinds involved and the AI-approvable change kinds.
Every limit-based reason names its config key and value.
Without a model call the comment says no model call, cost 0 USD.
General size ([size], docs not counted) → ORANGE, the model still runs; model limit → YELLOW, derived from [budget] max_usd_per_review and the new [model] max_context_tokens (replaces max_input_tokens).

Details: the PR document under docs/prs/.

🤖 Generated with Claude Code

werkrevisor 0.7.0: clearer reasons and two size limits. - The reason for change kind `mixed` names the file kinds involved and the AI-approvable change kinds. - Every limit-based reason names its config key and value. - Without a model call the comment says `no model call, cost 0 USD`. - General size (`[size]`, docs not counted) → ORANGE, the model still runs; model limit → YELLOW, derived from `[budget] max_usd_per_review` and the new `[model] max_context_tokens` (replaces `max_input_tokens`). Details: the PR document under `docs/prs/`. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-code added 6 commits 2026-10-05 15:27:29 +02:00

The reason for a change kind that is not AI-approvable names the file kinds involved and the approvable change kinds 5943c937b6

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

Every limit-based reason names the config key and value it comes from d27d4a1ece

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

A run without a model call reports cost 0 USD instead of an unknown cost 67c2b82741

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

Two size limits: a large PR is ORANGE but still reviewed; the model input limit derives from the budget and max_context_tokens f55fa74bd5

[size] counts only non-doc files and gives ORANGE without skipping the model.
[model] max_input_tokens becomes max_context_tokens, a quality cap;
the effective input limit is the lower of it and the limit the budget can pay for.
Example configs: 800 or 1000 changed lines, 0.40 USD per review for werkprobe and werkrevisor.

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

werkrevisor 0.7.0: README and German guides describe the two size limits and the new reason texts 20dd63e553

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

docs: PR document for clearer reasons and two size limits

werkator build succeeded after 00:05

Details

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

Details

ai-review/gate build succeeded after 00:01

Details

36cc6e3acf

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

claude-code marked the pull request as ready for review 2026-10-05 15:27:40 +02:00

ai-reviewer commented

2026-10-05 15:27:52 +02:00

🟡 werkrevisor: YELLOW / HUMAN_REVIEW

Change kind: mixed.

Reasons

YELLOW test changes are not purely additive: tests/test_classifier.py:144 removes a test or assertion, tests/test_classifier.py:145 removes a test or assertion, tests/test_classifier.py:146 removes a test or assertion, tests/test_classifier.py:147 removes a test or assertion, tests/test_classifier.py:149 removes a test or assertion and 12 more
YELLOW change kind 'mixed' is not AI-approvable
YELLOW risky paths changed: config/hsadmin-ng.toml (config/), config/werkprobe.toml (config/), config/werkrevisor.toml (config/**), werkrevisor/config.py (werkrevisor/config.py)
YELLOW full-file context incomplete: README.md (34810 chars > max_full_file_chars)
YELLOW full-file context incomplete: docs/anleitung-dev.hostsharing.net.md (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: docs/betrieb-git.javagil.de.md (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: docs/prs/2026-10-05-PR#4-clearer-reasons-and-two-size-limits.md (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_classifier.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_config_budget_redact.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_flow.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_fullfiles.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_pr_agent_integration.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_pr_agent_run.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_rules.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: tests/test_verdict.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/budget.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/classifier.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/cli.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/config.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/fullfiles.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/report.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/review.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/rules.py (total would exceed max_full_files_chars)
YELLOW full-file context incomplete: werkrevisor/verdict.py (total would exceed max_full_files_chars)
YELLOW too large for automatic review: about 47878 input tokens exceed the limit of 25000

_{Usage: model `openrouter/anthropic/claude-sonnet-5.5`, LLM skipped (too large), duration 0.9 s, cost unknown.

Reviewed head `36cc6e3acf0f98559729e3e0e66aabd885dcf84a` against base `6856d04398c32a56590be178b4b24fbe7cc75cc3`, werkrevisor 0.6.0.}

## 🟡 werkrevisor: YELLOW / HUMAN_REVIEW Change kind: `mixed`. ### Reasons - **YELLOW** test changes are not purely additive: tests/test_classifier.py:144 removes a test or assertion, tests/test_classifier.py:145 removes a test or assertion, tests/test_classifier.py:146 removes a test or assertion, tests/test_classifier.py:147 removes a test or assertion, tests/test_classifier.py:149 removes a test or assertion and 12 more - **YELLOW** change kind 'mixed' is not AI-approvable - **YELLOW** risky paths changed: config/hsadmin-ng.toml (config/**), config/werkprobe.toml (config/**), config/werkrevisor.toml (config/**), werkrevisor/config.py (werkrevisor/config.py) - **YELLOW** full-file context incomplete: README.md (34810 chars > max_full_file_chars) - **YELLOW** full-file context incomplete: docs/anleitung-dev.hostsharing.net.md (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: docs/betrieb-git.javagil.de.md (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: docs/prs/2026-10-05-PR#4-clearer-reasons-and-two-size-limits.md (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_classifier.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_config_budget_redact.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_flow.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_fullfiles.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_pr_agent_integration.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_pr_agent_run.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_rules.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: tests/test_verdict.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/budget.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/classifier.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/cli.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/config.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/fullfiles.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/report.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/review.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/rules.py (total would exceed max_full_files_chars) - **YELLOW** full-file context incomplete: werkrevisor/verdict.py (total would exceed max_full_files_chars) - **YELLOW** too large for automatic review: about 47878 input tokens exceed the limit of 25000 <sub> Usage: model `openrouter/anthropic/claude-sonnet-5.5`, LLM skipped (too large), duration 0.9 s, cost unknown.<br> Reviewed head `36cc6e3acf0f98559729e3e0e66aabd885dcf84a` against base `6856d04398c32a56590be178b4b24fbe7cc75cc3`, werkrevisor 0.6.0. </sub>

ai-reviewer requested review from mi 2026-10-05 15:27:52 +02:00

mi merged commit 3d174cc8a5 into main

2026-10-05 15:37:18 +02:00

mi deleted branch claude-code/clearer-reasons-and-size-limits

2026-10-05 15:37:18 +02:00

Sign in to join this conversation.

2 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: mi/werkrevisor#4