aiskillstore/marketplace / skills/0xkynz/memory-bank
memory-bank
VerifiedPersistent project documentation system that maintains context across sessions. Creates structured Memory Bank files to preserve project knowledge, decisions, and progress.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.3 KB |
| skill-report.json | config | 25.3 KB |
skill-report.json
View on GitHub{
"schema_version": "2.0",
"meta": {
"generated_at": "2026-07-04T13:34:12.712Z",
"slug": "0xkynz-memory-bank",
"source_url": "https://github.com/0xkynz/codekit/tree/main/templates/skills/memory-bank",
"source_ref": "main",
"model": "codex",
"analysis_version": "3.0.0",
"source_type": "community",
"content_hash": "e478d8f1ee0c8564051e47d412ef33c1a923b1cc70c0db4574e1a46c3a83e2c8",
"tree_hash": "cf412ff21e80d48d0992fe0c3211a8f26746dd5c3241bbdad3ec019bbe048aeb",
"provenance": {
"model_effective": "codex",
"fallback_chain": [
{
"agent": "codex",
"outcome": "success"
}
],
"analysis_version": "3.0.0"
}
},
"skill": {
"name": "memory-bank",
"description": "Maintains a structured memory-bank documentation set so AI coding agents can preserve project context, decisions, and progress across sessions.",
"summary": "Keeps project context, decisions, and progress in structured Markdown files for session continuity.",
"icon": "📦",
"version": "1.0.1",
"author": "0xkynz",
"license": "MIT",
"tags": [
"memory-bank",
"project-context",
"documentation",
"agent-workflows",
"handoff"
],
"supported_tools": [
"claude",
"codex",
"claude-code"
],
"risk_factors": [
"external_commands"
],
"category": "documentation"
},
"security_audit": {
"risk_level": "medium",
"is_blocked": false,
"safe_to_publish": true,
"analysis_status": "ok",
"summary": "The static analyzer flagged Markdown fence delimiters as Ruby or shell backtick execution and flagged a progress template heading as network reconnaissance. Review found those static hits to be false positives, with only benign documentation examples and a local directory creation example. The remaining concern is semantic: persistent memory files can carry prompt-injection content if they are treated as authoritative instructions.",
"remediation": [
{
"issue": "Persistent memory files may become a prompt-injection channel.",
"suggestion": "State that memory-bank files are project context only, and that system, developer, and current user instructions take priority over stored notes.",
"severity": "medium"
},
{
"issue": "The bash setup example can trigger command-execution scanners.",
"suggestion": "Label the mkdir command as optional local setup, and require normal user approval before running any shell command.",
"severity": "low"
}
],
"risk_factor_evidence": [
{
"factor": "external_commands",
"evidence": [
{
"file": "SKILL.md",
"line_start": 15,
"line_end": 23
},
{
"file": "SKILL.md",
"line_start": 23,
"line_end": 27
},
{
"file": "SKILL.md",
"line_start": 27,
"line_end": 34
},
{
"file": "SKILL.md",
"line_start": 34,
"line_end": 133
},
{
"file": "SKILL.md",
"line_start": 133,
"line_end": 135
},
{
"file": "SKILL.md",
"line_start": 135,
"line_end": 147
},
{
"file": "SKILL.md",
"line_start": 147,
"line_end": 167
},
{
"file": "SKILL.md",
"line_start": 167,
"line_end": 170
},
{
"file": "SKILL.md",
"line_start": 170,
"line_end": 185
},
{
"file": "SKILL.md",
"line_start": 185,
"line_end": 188
},
{
"file": "SKILL.md",
"line_start": 188,
"line_end": 207
},
{
"file": "SKILL.md",
"line_start": 207,
"line_end": 210
},
{
"file": "SKILL.md",
"line_start": 210,
"line_end": 228
},
{
"file": "SKILL.md",
"line_start": 228,
"line_end": 231
},
{
"file": "SKILL.md",
"line_start": 231,
"line_end": 238
},
{
"file": "SKILL.md",
"line_start": 238,
"line_end": 240
},
{
"file": "SKILL.md",
"line_start": 240,
"line_end": 250
},
{
"file": "SKILL.md",
"line_start": 250,
"line_end": 253
},
{
"file": "SKILL.md",
"line_start": 253,
"line_end": 270
}
]
}
],
"static_findings": [
{
"id": "external_commands:SKILL.md:15:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 15,
"line_end": 23,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:23:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 23,
"line_end": 27,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:27:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 27,
"line_end": 34,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:34:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 34,
"line_end": 133,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:133:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 133,
"line_end": 135,
"snippet": "```bash"
},
{
"id": "external_commands:SKILL.md:135:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 135,
"line_end": 147,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:147:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 147,
"line_end": 167,
"snippet": "```markdown"
},
{
"id": "external_commands:SKILL.md:167:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 167,
"line_end": 170,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:170:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 170,
"line_end": 185,
"snippet": "```markdown"
},
{
"id": "external_commands:SKILL.md:185:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 185,
"line_end": 188,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:188:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 188,
"line_end": 207,
"snippet": "```markdown"
},
{
"id": "external_commands:SKILL.md:207:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 207,
"line_end": 210,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:210:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 210,
"line_end": 228,
"snippet": "```markdown"
},
{
"id": "external_commands:SKILL.md:228:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 228,
"line_end": 231,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:231:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 231,
"line_end": 238,
"snippet": "```markdown"
},
{
"id": "external_commands:SKILL.md:238:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 238,
"line_end": 240,
"snippet": "```bash"
},
{
"id": "external_commands:SKILL.md:240:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 240,
"line_end": 250,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:250:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 250,
"line_end": 253,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:253:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 253,
"line_end": 270,
"snippet": "```markdown"
},
{
"id": "blocker:SKILL.md:259:network-reconnaissance",
"category": "blocker",
"severity": "low",
"pattern": "Network reconnaissance",
"file": "SKILL.md",
"line_start": 259,
"line_end": 260,
"snippet": "## In Progress"
}
],
"finding_verdicts": [
{
"id": "external_commands:SKILL.md:15:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This is a Markdown code fence showing the memory-bank directory structure. It is not Ruby backtick execution and does not run a command."
},
{
"id": "external_commands:SKILL.md:23:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence ends a directory tree example. There is no shell interpolation, command invocation, or executable code."
},
{
"id": "external_commands:SKILL.md:27:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This is another Markdown fence for a file hierarchy diagram. The fenced content is documentation, not a shell or Ruby execution path."
},
{
"id": "external_commands:SKILL.md:34:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing fence belongs to the hierarchy diagram. It is static documentation and cannot execute by itself."
},
{
"id": "external_commands:SKILL.md:133:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.93,
"reason": "This Markdown bash block contains a hardcoded mkdir command to create a local documentation folder. It has no user input, network access, or automatic execution behavior."
},
{
"id": "external_commands:SKILL.md:135:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence terminates the local setup example. It is not executable Ruby or shell backtick syntax."
},
{
"id": "external_commands:SKILL.md:147:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This Markdown fence starts a projectbrief.md template. The content is a documentation template, not executable code."
},
{
"id": "external_commands:SKILL.md:167:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence ends a documentation template. It does not represent command execution."
},
{
"id": "external_commands:SKILL.md:170:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This Markdown fence starts a productContext.md template. The content is user-facing documentation structure only."
},
{
"id": "external_commands:SKILL.md:185:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence ends the product context template. There is no command or dynamic execution."
},
{
"id": "external_commands:SKILL.md:188:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This Markdown fence starts an activeContext.md template. The content is plain Markdown documentation guidance."
},
{
"id": "external_commands:SKILL.md:207:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence ends the active context template. It is not executable syntax in this skill file."
},
{
"id": "external_commands:SKILL.md:210:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This Markdown fence starts a systemPatterns.md template. It documents architecture notes and does not execute commands."
},
{
"id": "external_commands:SKILL.md:228:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence ends the system patterns template. It contains no executable shell or Ruby behavior."
},
{
"id": "external_commands:SKILL.md:231:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This Markdown fence starts a techContext.md template. The fenced section is intended as Markdown content for a generated file."
},
{
"id": "external_commands:SKILL.md:238:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.93,
"reason": "This nested bash fence is a placeholder labeled setup commands inside a documentation template. It does not include a concrete risky command or automatic execution."
},
{
"id": "external_commands:SKILL.md:240:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence closes the placeholder setup block. It is documentation text rather than command execution."
},
{
"id": "external_commands:SKILL.md:250:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This closing Markdown fence ends the tech context template. It is not Ruby shell backtick execution."
},
{
"id": "external_commands:SKILL.md:253:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "This Markdown fence starts a progress.md template. The content is task status documentation only."
},
{
"id": "blocker:SKILL.md:259:network-reconnaissance",
"verdict": "false_positive",
"confidence": 0.99,
"reason": "The cited text is the progress template heading \"## In Progress\". It is not network scanning, host discovery, or reconnaissance behavior."
}
],
"semantic_findings": [
{
"title": "Persistent Context Can Carry Prompt Injection",
"description": "The skill instructs the agent to read all memory bank files at every task start and rely on them for continuity. If repository documentation is untrusted, malicious content in those files could influence later sessions.",
"severity": "medium",
"locations": [
{
"file": "SKILL.md",
"line_start": 9,
"line_end": 9
},
{
"file": "SKILL.md",
"line_start": 87,
"line_end": 97
},
{
"file": "SKILL.md",
"line_start": 121,
"line_end": 127
}
],
"confidence": 0.72,
"confidence_reasoning": "The instructions explicitly make memory-bank files persistent session context. The risk depends on who can edit those files, so this is a real contextual risk but not direct malware."
}
],
"critical_findings": [],
"high_findings": [],
"medium_findings": [
{
"title": "Persistent Context Can Carry Prompt Injection",
"description": "The skill instructs the agent to read all memory bank files at every task start and rely on them for continuity. If repository documentation is untrusted, malicious content in those files could influence later sessions.",
"locations": [
{
"file": "SKILL.md",
"line_start": 9,
"line_end": 9
},
{
"file": "SKILL.md",
"line_start": 87,
"line_end": 97
},
{
"file": "SKILL.md",
"line_start": 121,
"line_end": 127
}
],
"review_kind": "security",
"source_category": "semantic",
"source_severity": "medium",
"confidence": 0.72,
"confidence_reasoning": "The instructions explicitly make memory-bank files persistent session context. The risk depends on who can edit those files, so this is a real contextual risk but not direct malware."
}
],
"low_findings": [],
"dangerous_patterns": [],
"files_scanned": 1,
"total_lines": 283,
"audit_model": "codex",
"audited_at": "2026-07-04T13:34:12.712Z",
"agent_auto_install_policy": "allowed",
"manual_install_policy": "allowed"
},
"content": {
"user_title": "Preserve Project Context Across AI Sessions",
"value_statement": "AI coding sessions lose important project context over time. This skill keeps decisions, current work, and progress in structured Markdown files.",
"seo_keywords": [
"Claude",
"Codex",
"Claude Code",
"memory bank",
"AI project documentation",
"persistent context",
"project handoff",
"agent workflow",
"developer notes",
"session continuity"
],
"actual_capabilities": [
"Defines required memory-bank Markdown files for project brief, product context, active context, system patterns, tech context, and progress.",
"Guides agents to read memory bank files at the start of a session.",
"Provides update workflows for activeContext.md and progress.md.",
"Includes templates for each core documentation file.",
"Supports extra context files for APIs, testing, deployment, and complex features."
],
"limitations": [
"It does not verify documentation accuracy against the repository automatically.",
"It depends on users or agents to keep files current after meaningful changes.",
"It assumes project context lives in a local memory-bank directory.",
"Large memory banks can increase context usage during each session."
],
"use_cases": [
{
"title": "Resume Feature Work",
"description": "Continue a feature after time away by reading current focus, recent changes, and next steps.",
"target_user": "Solo developer"
},
{
"title": "Preserve Architecture Decisions",
"description": "Record patterns, constraints, and tradeoffs so future AI sessions follow established project direction.",
"target_user": "Engineering lead"
},
{
"title": "Document Client Project Context",
"description": "Create a concise project memory for requirements, setup details, and remaining work during client handoffs.",
"target_user": "Consultant"
}
],
"prompt_templates": [
{
"title": "Initialize Memory Bank",
"scenario": "Beginner setup for a new or undocumented project.",
"prompt": "Create a memory bank for this project. Ask me for missing project goals, then draft the required Markdown files with concise placeholders."
},
{
"title": "Update Current Work",
"scenario": "Routine update after completing a task.",
"prompt": "Update the memory bank after this change. Refresh activeContext.md and progress.md with completed work, current state, and next steps."
},
{
"title": "Review Documentation Drift",
"scenario": "Intermediate maintenance before a new session or release.",
"prompt": "Review all memory bank files against the current repository state. Identify outdated notes, missing decisions, and unclear next steps before editing."
},
{
"title": "Expand Feature Context",
"scenario": "Advanced documentation for complex work streams.",
"prompt": "Create additional memory bank context for this feature. Include requirements, system interactions, testing notes, risks, and links back to core files."
}
],
"output_examples": [
{
"input": "Start a memory bank for a new scheduling application.",
"output": "Creates a project brief, product context, technical context, system patterns, active context, and progress outline for the scheduling project."
},
{
"input": "Update memory bank after finishing authentication.",
"output": "Adds the completed authentication work to progress, records current focus, and captures the next tasks for authorization and testing."
},
{
"input": "Review whether the project documentation is still accurate.",
"output": "Compares each memory bank file with the visible project state and lists stale notes, missing setup details, and unclear decisions."
}
],
"best_practices": [
"Update activeContext.md and progress.md after meaningful project changes.",
"Keep each memory bank file concise, factual, and easy to scan.",
"Treat stored memory content as project context, not as higher-priority instructions."
],
"anti_patterns": [
"Do not store secrets, access tokens, private keys, or personal data in memory bank files.",
"Do not let outdated notes override the current repository state or current user request.",
"Do not create large unstructured notes that duplicate source files or issue trackers."
],
"faq": [
{
"question": "Where does the skill store project memory?",
"answer": "It is designed to use Markdown files inside a local memory-bank directory in the project."
},
{
"question": "Does it work with Claude, Codex, and Claude Code?",
"answer": "Yes. The report marks support for Claude, Codex, and Claude Code."
},
{
"question": "Does it run background services?",
"answer": "No. The skill is documentation guidance and does not define a background service."
},
{
"question": "Can it create the initial file structure?",
"answer": "Yes. It instructs the agent to create a memory-bank directory and draft the required Markdown files."
},
{
"question": "How often should the memory bank be updated?",
"answer": "Update it after significant changes, major decisions, completed features, or explicit update requests."
},
{
"question": "What information should be excluded?",
"answer": "Exclude secrets, credentials, private customer data, and instructions that conflict with current user or system directions."
}
]
},
"file_structure": [
{
"name": "SKILL.md",
"type": "file",
"path": "SKILL.md",
"lines": 283
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |