1
67%
main: 67%

Ran 02 Jun 2026 08:31AM UTC

Files 398

Run time 9s

Badge

Embed ▾

Committed 02 Jun 2026 08:26AM UTC coverage: 66.532% (-0.02%) from 66.548%

Job # 26807876112.1

Build Type

push

github

Committed by

web-flow

Commit Message

chore(config): Purge forbidden words and tune persona defaults (#720)

Replace all occurrences of "load-bearing", "comprehensive", and "honest
take" across persona and knowledge config files with clearer
alternatives ("structural", "foundational", "essential", "thorough",
etc.).

Extend the default persona's `pre_response_analysis` step to explicitly
instruct the model to scan its draft for forbidden words before
responding, and add "honest take" to the `forbidden_words` list.

Downgrade the reasoning effort from `xhigh` to `high` across all
personas (`architect`, `dev`, `pr-reviewer`, `pr-triager`,
`rfd-implementor`, `rfd-reviewer`, `rfd-triager`).

Additional housekeeping: add a `zai` model alias for
`cerebras/zai-glm-4.7`, tighten the `.ignore` glob patterns to be
explicit about crate subdirectory layouts, switch the
`project-structure` knowledge attachment from crate names to crate
paths, and expand the `cargo check` tool summary to mention doc comment
formatting.

Signed-off-by: Jean Mertz <git@jeanmertz.com>

Coverage Stats

33447 of 50272 relevant lines covered (66.53%)

250.1 hits per line

dcdpr / jp / 26807876112 / 1
67%
main: 67%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 26807876112.1

dcdpr / jp / 26807876112 / 1 67% main: 67%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 26807876112.1

dcdpr / jp / 26807876112 / 1
67%
main: 67%

README BADGES
x