vision-memory-mcp

A guide for using a Visual Memory MCP server to store webpage and application screenshots, their descriptions, and the transitions between interface states. MCP is a standard way for an agent to connect to external tools and services.

In plain words
What is it for?
Use it when capturing screenshots, checking visual states, navigating an interface, or saving and recalling visual milestones.
Why use it?
It helps the agent recognize screens it has already seen, avoid repeated image analysis, and keep track of how actions change the interface.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/putervision/webcrypt/vision-memory-mcp
Any agent
npx skills add putervision/WebCrypt --skill vision-memory-mcp
Clone the repo
git clone --depth 1 https://github.com/putervision/WebCrypt

Made for: Claude Code, Codex.

Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,449 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. Scan, not verified.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00034 $0.01449
Opus 5 $0.00017 $0.00724
Sonnet 5 $0.00007 $0.00290
Haiku 4.5 $0.00003 $0.00145

Measured 2d ago against content hash 2ab763b4643f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

vision-memory-mcp scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- **Google Antigravity (`~/.gemini/config/config.json`)**: Add these rules to your `"globalPermissionGrants"` -> `"allow"` list:
Origin

This is a copy

100% identical to vision-memory-mcp — 66 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.agents/skills/vision-memory-mcp/SKILL.md · 64 lines

How it starts

The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Visual Memory (vision-memory-mcp)

This project utilizes vision-memory-mcp to cache visual states, record layout transitions, and avoid repetitive LLM vision calls.

1. Priority Order & Checklist

Whenever you capture a screenshot, examine a webpage, or need to verify a visual state, you MUST run this sequence:

  1. Orient: Call get_session_context to load active transitions and recent visual states.
  2. Search (Optional): Call recall_memory to search past visual states by text query or image query.
  3. Ingest/Verify: Call analyze_screenshot with the base64 screenshot before calling any vision LLMs.
  4. Cache Hit: If is_known is true, read the returned description and do NOT call your vision LLM.
  5. Cache Miss: If is_known is false, inspect the image with your vision model, summarize the layout, and register it back by calling analyze_screenshot with both the screenshot and description parameters.
  6. Log Transitions: Right after taking any UI action (click, type, navigate, scroll), call record_outcome to build the navigation graph.
  7. Snapshotting: Call manage_snapshot (action: "save") when reaching milestones, and manage_snapshot (action: "diff") to check for visual regressions.

2. Complete Tool Reference

Tool Name Key Inputs Description
analyze_screenshot screenshot? (base64), file_path?, description?, items? Main ingestion (single or batch) and visual state retrieval tool.
recall_memory query?, screenshot?, file_path?, strategy?, limit? Search visual memory by text query or image query (read-only).
record_outcome from_state_id, to_state_id?, action, action_type? ('blocker' | 'click' | etc.) Record UI action transitions or log visual blockers for state-memory.
get_navigation_paths from_state_id?, to_state_id?, to_description?, max_hops? Find historical path or instructions between states.
predict_next_action current_state_id, goal_description?, goal_state_id? Predict best next UI action and grounded element handles (target_selector, target_coords).
compare_states state_a_id & state_b_id OR video_a_id & video_b_id Compare two states visually (has_layout_change) or compare video runs.
get_session_context include_recent?, include_frequent? Get recent/frequent states, transition graphs, disk stats, cache metrics, and version info.
manage_snapshot action ('save' | 'diff' | 'export' | 'restore'), name?, archive_json? Unified snapshot management for visual checkpoints and regression detection.
manage_visual_spec action ('set' | 'verify' | 'list'), name?, screenshot?, tolerance? Register and verify visual design contract baselines (Visual SDD).
manage_video action ('ingest' | 'search' | 'timeline'), file_path?, query?, video_id? Ingest WebM/MP4 recordings, search video keyframes, or retrieve timelines.
create_evidence_pack keyframe_state_ids, source_video_id?, linked_state_memory_nodes? Package immutable evidence packs linking video keyframes to state-memory DAGs.
export_trajectories format? ('json' | 'llava' | 'qwen2_vl' | 'joint'), trace_id? Export multimodal trajectories for model fine-tuning or joint workflow exports.
undo_visual_mutation type? ('state' | 'transition' | 'any') Revert the last visual state ingestion or transition edge addition.
forget_state state_id Purge a specific state and vector embedding for privacy.
wait_for_visual_state target_state_id, timeout_ms? Poll for target visual state until present or timeout occurs.

Read the full file on GitHub · 64 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 64 lines · 34 tokens per session scan B 2ab763b4643f

Subscribe to this mod's changes

vision-memory-mcp is a skill published in the GitHub repository putervision/WebCrypt (24 stars, last pushed 11d ago), licensed MIT. It adds 34 tokens to every session and 1,449 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). It is 100% identical to vision-memory-mcp, differing in 66 lines, and is treated as a copy.

Related

Other skills, from other repositories

Cryptographic Analysis & Assessment

SSL/TLS auditing, cipher suite analysis, hash algorithm identification, encryption implementation review, and cryptographic weakness detection in code.

Masriyan/Claude-Code-CyberSecurity-Skill · 30 tokens

analyzing-ransomware-encryption-behavior

Analyzes how a ransomware sample encrypts files: identifying the crypto scheme (symmetric/asymmetric/hybrid), key handling, file targeting and extension/marker changes, shadow-copy deletion, and ransom-note artifacts. Activates for requests to analyze ransomware encryption, assess decryptability, or study ransomware…

meltedinhex/analyst-ai-pack · 72 tokens

covalent-bond-test

How to write, run, and debug tests for Covalent Bond: the runner, the in-process mock relay, deterministic polling, port allocation, cleanup conventions, and what every kind of change must cover. Use when adding or modifying a test, when a suite fails, or before committing any behavior change.

gopalrajsuresh/covalent-bond · 68 tokens

covalent-bond-run

How to run Covalent Bond on a machine and pair two AI agents, covering how to register the MCP server, start a relay (mock or deployed), create/join a session, send files, and troubleshoot. Use when a user wants to set up, start, connect, or test Covalent Bond, or asks why a session won't connect.

gopalrajsuresh/covalent-bond · 77 tokens

covalent-bond-self-relay

Deploy the user's own Covalent Bond relay to their Cloudflare account: install wrangler, walk them through the browser login, run the deploy, verify /health, and set COVALENTRELAYURL. Use when a user asks to set up, deploy, or host their own relay instead of using the public one.

gopalrajsuresh/covalent-bond · 75 tokens

covalent-bond-guardrails

Guardrails for any agent editing the Covalent Bond repository, covering the security invariants that must never be broken, what code NOT to add, and the verify-don't-guess working discipline. Use whenever writing or changing code, tests, or config in this repo before making the edit.

gopalrajsuresh/covalent-bond · 68 tokens