← All Test Cases

SID-001

exact source id

critical

Repetitions

Documents

Questions

Reasoning

DIRECT

source-id exact-id no-approximation

📖 In Plain English

What this category tests

Does the brain return the exact KB-* source label, not an internal ID?

How the test works

After ingesting documents, the test asks the brain to cite specific sources. Citations must use the KB-* label format the document was stored with, never the internal src_* IDs.

Why it matters

Source IDs are the brain's contract for provenance. If they're wrong, evidence can't be traced.

⚙️ How a single rep runs

① Generate

Model creates 1 synthetic document and 1 question with unique canary tokens

→ Fresh content per run prevents memorization and proves real retrieval

② Ingest (MCP)

Model calls brain_ingest to store the 1 document

→ Tests the brain's storage and indexing pipeline

③ Query (MCP)

Model answers the question using brain retrieval tools (search, fetch, context_pack, etc.)

→ Core test — does the brain return correct evidence and let the model build a faithful answer?

④ Evaluate

Model judges the answer against ground truth (the document it generated in phase 1)

→ Produces a score 0–100 with detailed sub-scores (retrieval, fidelity, reasoning, etc.)

This rep is run 3 times per test run. A pass requires score ≥ 85 and no critical failures.

🔬 Technical Instructions (raw prompts sent to AI)

🔧 ① Setup Instructions 1264 chars

Generate one document containing a specific object with a unique code. Ingest it.
Generate a question asking for the exact source_id of the document containing that code.

Requirements:
- Invent a fictional object and a short alphanumeric code (e.g. "map R-19", "token AX7").
- Document content: "The <location> contains <object with code>." One sentence.
- source_id: KB-{{RUN}}-SID-001-A-v1
- Question: "What is the exact source_id of the document mentioning <code>? Return only the source_id and the exact sentence."

Output ONLY this JSON:
{
  "run_id": "{{RUN}}",
  "test_id": "SID-001",
  "repetition": {{REP}},
  "documents": [
    {"source_id": "KB-{{RUN}}-SID-001-A-v1", "content": "<sentence>", "title": "<title>", "version": 1}
  ],
  "questions": ["What is the exact source_id of the document mentioning <code>? Return the source_id and exact sentence."],
  "expected_answers": [{
    "question_index": 0,
    "correct_answer": "KB-{{RUN}}-SID-001-A-v1",
    "required_source_ids": ["KB-{{RUN}}-SID-001-A-v1"],
    "required_quote": "<exact sentence>",
    "required_label": "DIRECT",
    "must_refuse": false,
    "notes": "Source_id must be exact: KB-{{RUN}}-SID-001-A-v1. Any normalisation, truncation, or invented ID is a critical failure."
  }]
}

▶ ② Execution Instructions 119 chars

Use ONLY MCP Knowledge Brain retrieval tools.
Question: {{QUESTION}}
Return ONLY this JSON:
{{REQUIRED_ANSWER_SCHEMA}}

⚖ ③ Judge Instructions 339 chars

Ground truth:
{{SETUP_CONTEXT_JSON}}

Tested AI answer:
{{EXEC_ANSWER_JSON}}

Check: the source_id in source_ids_used and claims must be exactly "KB-{{RUN}}-SID-001-A-v1". No approximation, no truncation.
Critical failures: any source_id that differs even by one character, any invented ID.

Return ONLY this JSON:
{{JUDGE_RESULT_SCHEMA}}

Critical Failure Conditions

hallucinated_source_id
source_id_approximated_or_truncated
wrong_source_id_format

Recent Run History

3 runs

When	Run ID	Pass Rate	Avg Score	Reps
2026-05-24 13:08	20260524T130808Z-kqze	0%	65.0	0/1	View →
2026-05-24 12:41	20260524T124148Z-z2do	100%	100.0	1/1	View →
2026-05-24 11:37	20260524T113756Z-kduj	0%	33.0	0/1	View →

📄 Raw YAML cases/exact_source_id/SID-001.yaml

schema_version: "1.0"
test_id: "SID-001"
category: "exact_source_id"
severity: "critical"
repetitions: 3
reasoning_type: "DIRECT"
num_documents: 1
num_questions: 1
tags: ["source-id", "exact-id", "no-approximation"]

setup_instructions: |
  Generate one document containing a specific object with a unique code. Ingest it.
  Generate a question asking for the exact source_id of the document containing that code.

  Requirements:
  - Invent a fictional object and a short alphanumeric code (e.g. "map R-19", "token AX7").
  - Document content: "The <location> contains <object with code>." One sentence.
  - source_id: KB-{{RUN}}-SID-001-A-v1
  - Question: "What is the exact source_id of the document mentioning <code>? Return only the source_id and the exact sentence."

  Output ONLY this JSON:
  {
    "run_id": "{{RUN}}",
    "test_id": "SID-001",
    "repetition": {{REP}},
    "documents": [
      {"source_id": "KB-{{RUN}}-SID-001-A-v1", "content": "<sentence>", "title": "<title>", "version": 1}
    ],
    "questions": ["What is the exact source_id of the document mentioning <code>? Return the source_id and exact sentence."],
    "expected_answers": [{
      "question_index": 0,
      "correct_answer": "KB-{{RUN}}-SID-001-A-v1",
      "required_source_ids": ["KB-{{RUN}}-SID-001-A-v1"],
      "required_quote": "<exact sentence>",
      "required_label": "DIRECT",
      "must_refuse": false,
      "notes": "Source_id must be exact: KB-{{RUN}}-SID-001-A-v1. Any normalisation, truncation, or invented ID is a critical failure."
    }]
  }

execution_instructions: |
  Use ONLY MCP Knowledge Brain retrieval tools.
  Question: {{QUESTION}}
  Return ONLY this JSON:
  {{REQUIRED_ANSWER_SCHEMA}}

judge_instructions: |
  Ground truth:
  {{SETUP_CONTEXT_JSON}}

  Tested AI answer:
  {{EXEC_ANSWER_JSON}}

  Check: the source_id in source_ids_used and claims must be exactly "KB-{{RUN}}-SID-001-A-v1". No approximation, no truncation.
  Critical failures: any source_id that differs even by one character, any invented ID.

  Return ONLY this JSON:
  {{JUDGE_RESULT_SCHEMA}}

critical_failures:
  - "hallucinated_source_id"
  - "source_id_approximated_or_truncated"
  - "wrong_source_id_format"