BC qdrant-ingestion-best-practices
Use this skill whenever building, designing, or debugging a RAG pipeline using Qdrant as the vector store. Covers ingestion pipelines, chunking standards, metadata schema design, hybrid dense+sparse retrieval with RRF, access control patterns, embedding model selection (BGE-M3 for hybrid, text-embedding-3-small for dense-only), collection architecture, normalization, deduplication, idempotency, and operational standards. Triggers include: any mention of Qdrant, RAG pipeline, vector ingestion, chunking, embeddings, hybrid search, payload filters, or access-controlled retrieval.
As a process C 62/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 3
✓ No critical or high findings
Medium and low: 3
-
low Secrets in code
secret-high-entropy-tokenguides/02-metadata-schema.md:249High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)"parent_doc_id": "slac…900",
quoted -
low Risky intent
intent-offensive-securityguides/09-access-control.md:23Offensive-security / dual-use content (legitimate for authorised testing; review intended use)**2. Privilege escalation surface:** If an LLM can influence the values of access control fields (even indirectly, through prompt injection), and those fields are authoritative for access decisions, y
-
low Risky intent
intent-offensive-securityguides/09-access-control.md:216Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Storing `allowed_agents` in chunk payload | Re-ingestion required on every access change; privilege escalation risk | Store agent→scope mappings in orchestration layer config |
Files scanned: 13. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 2, column 14: description: Use this skill whenever building, designing, or debugging a RAG pi… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value
Process rating: all ten parameters 62/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 1 mutating operations with no state check
- 55Failures and branches. 1 branches
- 70When it triggers. States when to use, but not when not to
- 100Tools and files. No external tools needed
- 100Steps. 10 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1599 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +3Description length 583: enough signal without eating the budget
- +4Structure: 16 headings
- +3Step-by-step instructions: 10 items
- +4Has examples (4 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 74.