AC changshen-agent-memory
Cross-session memory recovery for AI agents — your agent never cold-starts again. Always-loaded identity & todos, hash-indexed conditional reads cut recovery cost ~70% (~55-75% cumulative when the host already injects identity). Pure local, zero deps. Use when: (1) A new chat asks "where did we leave off?" and you have no context (2) Context is filling up and you're about to hit the limit mid-task (3) User says "I already told you this" (4) You re-read the same identity/rules/todos files at every session start (5) Session start burns tens of thousands of tokens before real work begins (6) You need to hand a long task to a fresh session without losing progress 中文触发:新对话"接着上次" / 上下文快满 / 用户说"我说过了" / 每轮重读同样的规则待办 / 开场烧掉几万 token / 长任务交接
Cross-session memory recovery for AI agents — your agent never cold-starts again.
As a process C 53/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- For Hermes the description must be one sentence under 60 characters; move the conditions to a "When to Use" section.
- 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: 10. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-long-hermesdescription is 740 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period)
Process rating: all ten parameters 53/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
- 60Consistency. The Hermes dialect needs category and tags
- 70Execution cost. Instruction body is 4246 tokens
- 100Steps. 34 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 12 top-level sections: this looks like several domains in one skill
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
- -5TODO / placeholder text left in the skill
- -217 emoji in the instructions: noise for the model
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +3Description length 740: enough signal without eating the budget
- +4Structure: 15 headings
- +3Step-by-step instructions: 34 items
- +4Has examples (3 code blocks)
- +3All 2 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 80.