Unblocks docker-image-scan (red 5 runs straight on GHSA-f88m-g3jw-g9cj: next's nested sharp@0.34.5, deduped via an override).
Finishes the #1218 Sonnet 5 rollout: compliance-pr and compliance-swarm were falling through to compliancemaxx's sonnet-4-6 default; swedish-compliance-review.mjs budgeted max_tokens as if thinking were off (it is adaptive-by-default on Sonnet 5) and never checked stop_reason; pr-agent's token budgets were sized for 4.6's tokenizer and its hidden default OpenAI fallback list is now emptied explicitly.
Core build 7m43s -> 2m51s measured (parallel checks/build/test, unit suite sharded 4 ways). Docker publish moves off QEMU to native ARM runners with a digest-merge job, so tags apply only on success and latest never moves on failure.
40 actions pinned to immutable SHAs; adds zizmor (0 high after fixing persist-credentials on 7 checkouts and permissions on test-pg-real) and CodeQL (0 findings on first run).
Full details in the PR body.
* feat(agent): move every model call to Sonnet 5
Sonnet 5 is verified enabled on our Bedrock account already: a live probe of
eu.anthropic.claude-sonnet-5 in eu-north-1 answered normally, so no model-access
request was needed. The bare anthropic.claude-sonnet-5 is rejected (on-demand
throughput needs the cross-region inference profile), so the eu. prefix we
already use stays.
This is not a model-string swap. Sonnet 5 REJECTS the fixed thinking budget
outright: thinking {type:'enabled', budget_tokens} returns 400 "not supported
for this model. Use thinking.type.adaptive and output_config.effort". Every
chat intent set a budget, so the assistant would have failed on the first turn
after a bare ID change. Reasoning depth is now an effort level (STANDARD high,
DEEP xhigh), and max_tokens is explicit per tier rather than derived from a
budget that no longer exists.
display:'summarized' is load-bearing, not cosmetic. The default is 'omitted',
which still emits thinking blocks but with empty text. Measured on our own
account at xhigh effort: summarized returned ~1k characters of reasoning, the
default returned none. Without it the collapsible "Tänker ..." block in the
chat would have gone silently empty, which no mocked test would have caught.
Ceilings are raised (16k standard, 24k deep) because Sonnet 5's tokenizer
produces roughly 30% more tokens for the same text and max_tokens now caps
thinking and the visible reply together.
Also resolves the Opus 4.7 landmine recorded in the readiness doc: the composer
comment told ops to flip BEDROCK_OPUS_MODEL_ID to Opus 4.7, which would have
400d every thinking intent against the legacy budget shape. Both model
constants now point at Sonnet 5 and the stale instruction is gone.
Checked but deliberately unchanged: forced tool_choice in atom-selection. The
Sonnet 5 docs require thinking:{type:'disabled'} alongside a forced tool_choice
on Bedrock; probed against our account, the forced call succeeds without it, so
no change was made rather than adding a guard we cannot show is needed.
Other call sites moved too: invoice-inbox extraction, document extraction, the
compliance config, and the CI/CD workflows (pr-agent MODEL and MODEL_WEAK,
swedish-compliance-review, compliance-swarm).
Verified: 11315 tests pass, lint and tsc clean on every touched file, guards
pass.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix(agent): review triage: keep the no-thinking output ceiling, finish the model sweep
max_tokens now caps thinking and the visible reply together, so collapsing the
two tiers into one made every non-thinking intent inherit a 16000 ceiling where
it used to have 4096. Give it its own MAX_TOKENS_NO_THINKING instead, set to the
old 4096 scaled ~30% for Sonnet 5's tokenizer so the effective reply length is
unchanged rather than quietly cut.
scripts/swedish-compliance-review.mjs still fell back to Sonnet 4.6 when
REVIEW_MODEL was unset, so a manual run silently used the old model. The initial
sweep only covered .ts and .yml.
pr-agent's FALLBACK_MODELS listed the primary model as its own fallback, which is
not a fallback; dropped it and rewrote the surrounding comments, which still
described Opus 4.8 and a 200k window.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
The unquoted "compliance: full swarm" value contains a colon+space,
which YAML parses as the start of a nested mapping, causing GitHub
Actions to reject the whole file as invalid.
Signed-off-by: Jonas Flodén <jonas@floden.nu>
Em dashes (—) and en dashes (–) had spread across comments, docs, tests,
and a few UI strings, reading as AI-generated boilerplate rather than
house style. Replaced each with punctuation matching its context: colon
for explanatory clauses, comma for asides, plain hyphen for numeric/legal
ranges (e.g. "21-23§"), "to"/"till" for date ranges, parentheses for
paired-dash asides. messages/en.json and messages/sv.json were fixed by
hand together to keep sv/en in sync.
Left untouched where the dash is the functional subject rather than
decorative punctuation: date-range-parser.ts's separator regex,
charset-repair.ts's CP1252 byte-mapping table (and its test), the SIE
encoding mojibake docs, generic-csv.ts's minus-sign normalizer, the
agent system-prompt files that already instruct against em dashes, and
a golden iXBRL test fixture compared byte-for-byte.
Also fixes two bugs surfaced along the way: an off-by-one in
ApiKeysPanel's scope-label split (a leftover from an earlier partial
pass), and a charset-repair test that had lost the literal en-dash it
exists to verify.
Regenerated the agent atom seed migration (skills:generate) since 27
SKILL.md files changed. Added a CLAUDE.md rule against em/en dashes,
with an explicit carve-out for the functional-dash cases above.
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
* chore(compliance): upgrade to compliancemaxx@v2
Per-PR check switches from `mode: pr` (deterministic scanners) to
`mode: review` (LLM-only diff review, ~90s, ~$0.05/PR). The nightly
audit workflow continues to provide the deeper scanner-backed coverage,
so concrete-pattern findings (secrets, CVEs, IaC misconfigs) are still
caught — just within 24 hours instead of immediately.
Trade-off accepted: PR feedback becomes shorter and more reasoned
(LLM judgement) rather than longer and more pattern-matched (scanner
output). Easier to act on per-PR; deep scanner coverage retained nightly.
- compliance-pr.yml: @v1 mode:pr → @v2 (default mode: review)
- compliance-swarm.yml: @v1 mode:swarm → @v2 mode:audit
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* chore: re-trigger to pick up v2.0.1 action.yml fix
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* chore(compliance): consume erp-mafia/compliancemaxx@v1
Adds the consumer side of the new compliancemaxx orchestrator:
- .github/workflows/compliance-pr.yml — single-job advisory check
that calls `uses: erp-mafia/compliancemaxx@v1` on every PR.
Threshold + skill selection live in .compliance/config.yml.
`fail-on-findings: false` keeps it strictly advisory while findings
bed in; flip to true after triage.
- .compliance/config.yml — bootstrap config: all 5 skills enabled
(oss-license, owasp-asvs-v5, iso-27001-2022, soc2, gdpr),
severity_threshold_to_block=critical, no suppressions yet.
The orchestrator code, skill manifests, and reference docs live in the
new public repo: https://github.com/erp-mafia/compliancemaxx
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* chore: retrigger compliance check
* chore(compliance): add full-swarm workflow (nightly + dispatch + label)
Adds the LLM-powered deep_audit alongside the existing PR-mode advisory
check. Triggers:
- 02:00 UTC nightly cron
- workflow_dispatch (manual)
- PRs labeled `compliance:full-audit`
Authenticates to AWS Bedrock via the existing AWS_ACCESS_KEY_ID/SECRET
repo secrets (same path swedish-compliance-review.yml already uses).
`fail-on-findings: false` keeps it observational while findings bed in.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Jakob Wennberg