CtrlK
BlogDocsLog inGet started
Tessl Logo

markusdowne/memory-roundtrip-guard

Tests memory writes, confirms read-back accuracy, and validates retrieval success to ensure saved information can actually be recovered. Use when you need to verify memory was saved correctly, check if stored data can be retrieved, confirm a memory entry is discoverable, or escalate when saved information appears lost or corrupted. Covers write confirmation, read-back comparison, retrieval smoke testing, and failure escalation. Includes explicit untrusted-content/prompt-injection guardrails for third-party inputs.

92

1.19x

Quality

90%

Does it follow best practices?

Impact

97%

1.19x

Average score across 5 eval scenarios

Overview
Skills
Evals
Files

rubric.jsonevals/scenario-2/

{
  "context": "Tests whether the agent captures required metadata fields when writing to memory, immediately reads back the saved entry, validates structural integrity, and produces a correctly structured output report.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Timestamp in metadata",
      "description": "The memory entry is written with a timestamp metadata field",
      "max_score": 8
    },
    {
      "name": "Source in metadata",
      "description": "The memory entry is written with a source metadata field",
      "max_score": 8
    },
    {
      "name": "Confidence in metadata",
      "description": "The memory entry is written with a confidence metadata field",
      "max_score": 8
    },
    {
      "name": "Content field present",
      "description": "The memory entry includes a content (value) field",
      "max_score": 6
    },
    {
      "name": "Write success check",
      "description": "The code or report explicitly checks/verifies the write operation succeeded (not assumed)",
      "max_score": 10
    },
    {
      "name": "Read-back performed",
      "description": "The code reads back the entry immediately after writing (a memory read/get is called for the same key)",
      "max_score": 12
    },
    {
      "name": "Metadata presence validated",
      "description": "The read-back is followed by a check that required metadata fields are present",
      "max_score": 10
    },
    {
      "name": "Content non-empty validated",
      "description": "The read-back validation includes a check that content/value is non-empty",
      "max_score": 8
    },
    {
      "name": "Write status in output",
      "description": "The output report/log includes a write status section or field",
      "max_score": 8
    },
    {
      "name": "Read-back status in output",
      "description": "The output report/log includes a read-back status section or field",
      "max_score": 8
    },
    {
      "name": "Classification in output",
      "description": "The output report/log includes a classification result (e.g., 'clean', 'operational', or 'critical')",
      "max_score": 7
    },
    {
      "name": "Remediation action in output",
      "description": "The output report/log includes a remediation or escalation action field",
      "max_score": 7
    }
  ]
}

Install with Tessl CLI

npx tessl i markusdowne/memory-roundtrip-guard@0.1.2

evals

SKILL.md

tile.json