implementer item-5 names tiny-real data-ingestion probe

Overview / Motivation

Auto-filed by the /daily Step-C parked-candidate routing pass (2026-07-09) from a recursion-guard-parked workflow-fix candidate raised on task #1157.

Goal

Add the #1092 data-ingestion probe class ('incl. the bounded tiny-real streaming probe for real-corpus ingestion phases') to experiment-implementer.md item 5's tiny-real bullet, so the evidence Step 6d.0-bis demands is also named on the producer side.

Workflow gap

• Bug observed: experiment-implementer.md item 5 tiny-real bullet (~lines 197-205) names only the #906 mock-seam recipe; #1157 named the data-ingestion tiny-real probe on the 6d.0-bis consumer/gate side but the producer-side bullet still omits it (verified on main 2026-07-09).
• Why it is a workflow gap: The smoke gate (6d.0-bis) now refuses seam-stubbed evidence for ingestion phases, but the agent spec that authors the smoke evidence never tells the implementer to produce the streaming probe.
• Confidence (emitter): medium

Proposed change (candidate diff sketch — refine in planning)

(none — one-clause addition to the item-5 tiny-real bullet naming the bounded tiny-real streaming probe for real-corpus ingestion phases, cf. #1092.)

Scope / surfaces

• Primary target: .claude/agents/experiment-implementer.md

Constraints / invariants

• Workflow-surface only. scripts/workflow_lint.py --check-asks passes; ruff on touched files passes.
• This session runs under EPM_WORKFLOW_FIX_SESSION=1 semantics (workflow_fix_target Provenance line) — it MUST NOT auto-route its own subagents' workflow-fix candidates.

Provenance

• workflow_fix_target: .claude/agents/experiment-implementer.md
• fingerprint: e35f0013608a

Parked prose-followup on #1157, 2026-07-09T09:53:56Z (Alternatives critic, Phase 2): target .claude/agents/experiment-implementer.md item 5 tiny-real bullet; add the #1092 data-ingestion probe class so the evidence 6d.0-bis demands is also named on the producer side. confidence: medium.