gemini(w5): preview projection maxTasks boundary test #126

Closed
opened 2026-05-28 01:20:43 +02:00 by codex · 1 comment
Collaborator

Parent: #2
Agent lane: Gemini 3.5 Flash
Wave: 5 / Swarmheart / Leviathan
Risk class: low

Goal

Ensure preview enforces safety caps before expensive or misleading work.

Context refs

  • docs/leviathan-projection-contract.md
  • packages/api/src/routers/agent.ts

Scope

  • Find preview projection implementation.
  • Add test around maxTasks exactly at, below, and above limit.
  • Assert errors are structured and non-mutating.

Acceptance

  • Too many tasks fail before preview construction.
  • No audit/write side effects occur in preview-only paths.
  • Error tells caller how to split batches.

Suggested checks

  • Projection preview tests.

Non-goals / fences

  • Do not deploy, restart production, rotate secrets, or run production migrations.
  • Do not widen MCP write authority or public exposure.
  • Keep the change small enough for one focused PR or one scouting report.

Expected output

A short PR or issue comment with findings, touched files, tests run, and remaining risks.

Parent: #2 Agent lane: Gemini 3.5 Flash Wave: 5 / Swarmheart / Leviathan Risk class: low ## Goal Ensure preview enforces safety caps before expensive or misleading work. ## Context refs - `docs/leviathan-projection-contract.md` - `packages/api/src/routers/agent.ts` ## Scope - Find preview projection implementation. - Add test around `maxTasks` exactly at, below, and above limit. - Assert errors are structured and non-mutating. ## Acceptance - Too many tasks fail before preview construction. - No audit/write side effects occur in preview-only paths. - Error tells caller how to split batches. ## Suggested checks - Projection preview tests. ## Non-goals / fences - Do not deploy, restart production, rotate secrets, or run production migrations. - Do not widen MCP write authority or public exposure. - Keep the change small enough for one focused PR or one scouting report. ## Expected output A short PR or issue comment with findings, touched files, tests run, and remaining risks.
Author
Collaborator

Implemented as PR #146.

Scope:

  • previewLeviathanProjection now accepts and enforces maxTasks before preview construction.
  • Shared API guard is used by preview/apply paths.
  • MCP preview_leviathan_projection exposes the same maxTasks input.
  • Integration test covers below-limit, at-limit, and above-limit previews.
  • Above-limit preview asserts no card creation and no agent audit side effects.

Checks run:

  • pnpm --filter @kan/api exec vitest integration-tests/agent.integration.test.ts --run -t "enforces Leviathan preview maxTasks"
  • pnpm --filter @kan/api exec vitest integration-tests/agent.integration.test.ts --run
  • pnpm --filter @kan/api typecheck
  • pnpm --filter @kan/mcp typecheck
  • pnpm --filter @kan/mcp test
  • pnpm exec prettier --check packages/api/src/routers/agent.ts packages/api/integration-tests/agent.integration.test.ts packages/mcp/src/client.ts packages/mcp/src/index.ts

No deploy, migration, token/scope change, or real projection write.

Implemented as PR #146. Scope: - `previewLeviathanProjection` now accepts and enforces `maxTasks` before preview construction. - Shared API guard is used by preview/apply paths. - MCP `preview_leviathan_projection` exposes the same `maxTasks` input. - Integration test covers below-limit, at-limit, and above-limit previews. - Above-limit preview asserts no card creation and no agent audit side effects. Checks run: - `pnpm --filter @kan/api exec vitest integration-tests/agent.integration.test.ts --run -t "enforces Leviathan preview maxTasks"` - `pnpm --filter @kan/api exec vitest integration-tests/agent.integration.test.ts --run` - `pnpm --filter @kan/api typecheck` - `pnpm --filter @kan/mcp typecheck` - `pnpm --filter @kan/mcp test` - `pnpm exec prettier --check packages/api/src/routers/agent.ts packages/api/integration-tests/agent.integration.test.ts packages/mcp/src/client.ts packages/mcp/src/index.ts` No deploy, migration, token/scope change, or real projection write.
Sign in to join this conversation.
No labels
3plus3-followup
agent/claude-code
agent/codex
agent/hermes
agent/iskra
agent/ollama
agent/patchwarden
analytics
api
cockpit
dependency/blocked
dependency/blocks-others
dependency/cross-repo
dependency/needs-confirmation
docs
domain:agents
domain:ci
domain:docs
domain:forgejo
domain:infra
domain:memory
domain:runtime
domain:signal
domain:ux
flow/architecture
flow/blocked
flow/deployed
flow/done
flow/implementation
flow/intake
flow/maintained
flow/observed
flow/ready
flow/refining
flow/retired
flow/review
gemini-flash
judge/codex-candidate
judge/hermes-candidate
judge/low-confidence
judge/needs-refinement
judge/operator-needed
judge/p0
judge/p1
judge/p2
judge/p3
judge/park
judge/patchwarden-candidate
judge/stale-priority
kind/adr
kind/bug
kind/chore
kind/feature
kind/infra
kind/ops
kind/refactor
kind/research
leviathan
mcp
merge/auto
merge/manual
merge/manual-dependency-conflict
merge/manual-failing-tests
merge/manual-merge-conflict
merge/manual-missing-review
merge/manual-operator-preference
merge/manual-red-zone
merge/manual-security-sensitive
merge/manual-unclear-scope
merge/manual-unknown
mode:operator-only
mode:patchwarden-iskra-approved
mode:safe-auto
observed/erroring
observed/needs-followup
observed/pending
observed/retire-candidate
observed/unused
observed/used
ops
priority:p0
priority:p1
priority:p2
priority:p3
review:claude-reviewed
review:codex-reviewed
review:dziadek-reviewed
review:needs-human
safety
safety:external-write
safety:no-prod-mutation
safety:prod-impact
safety:secret-touch
scout
security
size/large
size/medium
size/small
size/tiny
size/unknown
small-task
source/adr
source/agent-generated
source/manual
source/operator-chat
source/voice-note
status:blocked
status:codex-ready
status:merged:pending-evidence
status:needs-evidence
status:operator-needed
status:parked
tests
tier:0-platform-substrate
tier:1-iskra-value-layer
tier:2-tools-products-modules
type:bug
type:chore
type:docs
type:feat
type:policy
type:research
ui
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
pdurlej/kan-ductor#126
No description provided.