Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →athola/memory-clarity-probe
Probe memory/summary clarity via dual anchor questions: task progress, info gaps.
Passed 21 of 22 test cases with the skill, and 7 of 22 without it.
and 11 more in the full report
14 fixed7 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
Before saving session-state.md and handing off to a continuation agent, evaluate this state file snippet: <memory> Completed setup of Kafka event consumer service. Consumer group order-processor active on topic orders-v1. Outstanding items: configure dead-letter queue topic orders-dlq and set max retry attempts to 5. </memory> Run the dual-probe clarity evaluation and report the exact outputs.
| Checked | Without skill | With skill |
|---|---|---|
| Progress probe rating is Clear | Missed | Passed |
| Gap probe rating is Bounded | Missed | Passed |
| Composite rating is Clear | Missed | Passed |
The instructions your model reads when the skill loads.
Assess whether a memory, summary, or session state retains enough task information to guide future reasoning.
A quality gate for any memory or summary, based on the dual-probe pattern from MMPO (arXiv:2605.30159, Liu et al. 2026). The probe asks two anchor questions against the current memory and evaluates whether the answers are confident and complete:
A clear memory answers the progress probe with specific, verifiable state (not vague placeholders) and enumerates bounded, concrete unknowns on the gap probe. An ambiguous memory produces hedging on the progress probe and open-ended uncertainty on the gap probe.
The two probes target different failure modes:
The MMPO paper's ablation (Table 4) shows progress+gap outperforms
gap-only across all context lengths. Use both probes.
This skill implements a qualitative clarity assessment. It does
test cases passed, vs 7 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.