AC compression-monitor
Detect behavioral drift in persistent AI agents after context compression events. Use when a long-running agent has compressed its context (compaction, truncation, or summarization) and you need to verify the agent is still behaving consistently. Measures three observable signals without requiring access to the agent's internals: ghost lexicon decay (loss of precise vocabulary), context consistency score (CCS via embedding similarity), and tool call distribution shift. Includes ready-to-use framework integrations for smolagents, Semantic Kernel, LangChain/DeepAgents, CAMEL, and the Anthropic Agent SDK. Triggers on: "context compression", "compaction", "agent drift", "behavioral drift", "ghost lexicon", "CCS", "context consistency", "did my agent change", "compression boundary", "long-running agent", "persistent agent drift", "context window rotation", or any task involving verifying agent behavioral consistency across session boundaries.
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: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
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 (compression-monitor) differs from the folder (morrow-compression-monitor)
- 70When it triggers. States when to use, but not when not to
- 100Tools and files. No external tools needed
- 100Steps. 10 steps
- 100Execution cost. Instruction body is 878 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)
- +3Description length 951: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -2localhost URLs: will not work for another user
- +1No license
- +2Single-language instructions
- +5Description quotes 11 example trigger phrases
- +4Structure: 9 headings
- +3Step-by-step instructions: 10 items
- +4Has examples (3 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.