analyse-agent-memory

Featured

Analyse memory and context routes as a specialist invoked by analyse-agentic-system, returning a typed report for integration into the main result.

AI & Automation 88 stars 11 forks Updated yesterday CC-BY-4.0

Install

View on GitHub

Quality Score: 89/100

Stars 20%
65
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Analyse agent memory Establish the system's memory mechanisms and their supported comparison classifications for integration into the main agentic-system analysis. ## Commission and boundary Run as a fresh specialist under `analyse-agentic-system`. Require the parent run ID, frozen `memory-input.md`, report destination, and permitted source access. The input supplies the subject, source register with full revision or capture digest and access root, relevant canonical records, requested memory scope and depth, exclusions, and any specific question. Its records are provisional findings to check against sources, not accepted conclusions. Write only the commissioned `memory-report.md` using `kb/reports/types/agent-memory-analysis-report.md`. Read that contract and the Memory comparison fields and Status fields sections of `kb/types/agentic-system-analysis-result.md`. Do not load the legacy review type, prior system reviews, surveys, matrix outputs, or style exemplars. The parent owns canonical IDs, integration, publication and completion. Do not publish, modify the parent's input/result, delegate, or stage and commit. Hash the input and this skill before analysis. Report those identities and your actual model identity; state `unknown` if the runtime does not expose it. Verify input and method hashes again before returning. Changed input requires a new handoff from the parent, not silent reconciliation in the worker. ## Inspect and explain Use the frozen primary sources. ...

Details

Author
zby
Repository
zby/commonplace
Created
6 months ago
Last Updated
yesterday
Language
Python
License
CC-BY-4.0

Similar Skills

Semantically similar based on skill content — not just same category