AC memory-palace
Long-term memory system for AI agents using a file-based "palace" architecture with BGE-M3 vector search, metadata filtering, compound scoring (semantic + recency + importance), GraphRAG Lite neighbor expansion, and memory metabolism protocols. Use for: (1) Building agentic long-term memory from scratch, (2) Upgrading flat file memory with vector search and temporal weighting, (3) Managing agent knowledge with automatic consolidation, cold zone archiving, and structured reflection. Trigger on phrases like "long-term memory", "vector search", "agent memory", "memory management", "knowledge base for agent".
As a process C 55/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
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 · 0
✓ No critical or high findings
Files scanned: 12. 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: Long-term memory system for AI agents using a file-based "palace" … ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value
Process rating: all ten parameters 55/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 40Consistency. Frontmatter name (memory-palace) differs from the folder (z1-memory-palace)
- 70When it triggers. States when to use, but not when not to
- 100Tools and files. No external tools needed
- 100Steps. 6 steps
- 100Execution cost. Instruction body is 611 tokens
- 100Running it twice. No mutating operations
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
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -32 of 6 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +5Description quotes 6 example trigger phrases
- +3Description length 612: enough signal without eating the budget
- +4Structure: 7 headings
- +3Step-by-step instructions: 6 items
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (1 of 4)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 80.