open/ms-ai-architect
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity Actions
198 commits 1 branch 8 tags 130 MiB
Commit graph

3 commits

Author SHA1 Message Date
Kjell Tore Guttormsen
f8a160585f test(ms-ai-architect): Sesjon 23 — auto_discover stale-felt assertion korrigert (søk-verifisert, 13/13) [skip-docs] 
Oppgave 0 (#44): søk-først (claude-code-guide mot offisiell docs) verifiserte at
`auto_discover` IKKE er et gyldig Claude Code plugin.json-felt — auto-discovery er
implisitt default (code.claude.com/docs/en/plugins-reference; ukjent felt gir kun
valideringsadvarsel, ikke feil). → stale konvensjon, IKKE manglende felt.

- tests/test-plugin-discovery.sh: stale hard-assertion (krevde `auto_discover: true`)
  → optional-felt-logikk som speiler validate-plugin.sh (fravær = PASS «auto-discovery
  implisitt»; present må være true). plugin.json URØRT (la IKKE til ugyldig felt).
  12/13 → 13/13.

validate 239/0 (ingen regresjon). STATE oppdatert: NESTE = C1 Tier 2.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-06-21 13:48:40 +02:00
Kjell Tore Guttormsen
039b61ca24 chore(ms-ai-architect): Sesjon 22 — rydd stale pre-edit-secrets-testreferanser [skip-docs] 
pre-edit-secrets.mjs-hooken ble bevisst fjernet i v1.7.0 (secrets scanning
consolidated to llm-security, CHANGELOG l.413), men 2 testfiler refererte den
fortsatt og ga falske feil.

- tests/test-hooks.sh: fjernet pre-edit-secrets-testblokk (5 tester) → 6/6 grønn.
- tests/test-plugin-discovery.sh: fjernet pre-edit-secrets fra CLAUDE.md-dok-loop.

Separat funn (IKKE fikset, scope-guard): test-plugin-discovery.sh har 1
gjenværende feil `auto_discover is not true` — plugin.json mangler feltet; ikke
hook-relatert, krever søk-først (gjeldende felt vs stale test). Flagget → task #44.

De 4 sporede suitene uendret (validate 239 · kb-update 155 · kb-eval 100 · kb-integrity 192/192).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-06-20 23:08:11 +02:00
Kjell Tore Guttormsen
baa2d0220b feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 
Add /ultraresearch-local for structured research combining local codebase
analysis with external knowledge via parallel agent swarms. Produces research
briefs with triangulation, confidence ratings, and source quality assessment.

New command: /ultraresearch-local with modes --quick, --local, --external, --fg.
New agents: research-orchestrator (opus), docs-researcher, community-researcher,
security-researcher, contrarian-researcher, gemini-bridge (all sonnet).
New template: research-brief-template.md.

Integration: --research flag in /ultraplan-local accepts pre-built research
briefs (up to 3), enriches the interview and exploration phases. Planning
orchestrator cross-references brief findings during synthesis.

Design principle: Context Engineering — right information to right agent at
right time. Research briefs are structured artifacts in the pipeline:
ultraresearch → brief → ultraplan --research → plan → ultraexecute.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-04-08 08:58:35 +02:00