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.
npx skills add mostafa-drz/claude-skills --skill screenshots-memorygit clone --depth 1 https://github.com/mostafa-drz/claude-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mostafa-drz/claude-skills/screenshots-memory)<a href="https://agentmods.dev/skills/mostafa-drz/claude-skills/screenshots-memory"><img src="https://agentmods.dev/badge/skills/mostafa-drz/claude-skills/screenshots-memory/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mostafa-drz/claude-skills/screenshots-memory"><img src="https://agentmods.dev/badge/skills/mostafa-drz/claude-skills/screenshots-memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00212 | $0.07138 |
| Opus 5 | $0.00106 | $0.03569 |
| Sonnet 5 | $0.00042 | $0.01428 |
| Haiku 4.5 | $0.00021 | $0.00714 |
Grade A, and why
screenshots-memory scanned grade A with 0 findings 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 today.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 503 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Screenshots Memory
Sync screenshots from anywhere on disk into a git-versioned semantic memory, read each one with Claude vision, extract kind-aware structured notes with real provenance, cluster them by topic, render clean HTML per cluster, answer questions over them, and get sharper at reading your particular apps every time you correct an extraction.
A screenshot is a snapshot of a moment you cared about. This skill stops those moments from rotting in a folder.
New here? README.md is the human-facing overview, and
examples/quickstart.html has copy-paste recipes for
everyday use.
Preferences
On startup, use Read to load ~/.claude/skills/screenshots-memory/preferences.md.
If it's missing, treat this as first-run (see First-time detection).
Defaults when no preferences exist:
memory-root:~/screenshots-memory/(confirmed on first run; git-init'd so every change is diffable and revertible)inbox:~/Desktop(where loose screenshots pile up; synced when no path is given)originals:move(move= once the store's copy is committed and byte-verified, the inbox original goes to the Trash; the store becomes the only copy outside it ·copy= inbox original left alone)confidence-threshold:0.75(extractions below this are flagged forreview, never silently trusted)cluster-style:topic(subject-based;kindis a filter chip on the page, not a folder)sensitive-policy:ask-once(flag during extraction; ask about all of them in a single round once every capture has been read)open-html:true(auto-open the cluster HTML when a sync finishes)tone:friendly-cli(terse, warm, direct)
The extraction guide, kind registry and corrections log live in the memory store,
not the skill dir — so they're versioned with the user's data and survive reset. On
startup, load {memory-root}/extraction-guide.md if it exists and treat it as ground
truth about this user's screen: which apps they use, that "ABI" is a client and not
a typo. Feeding it to the extractor every sync is what makes the memory improve.
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today First seen · 503 lines · 212 tokens per session scan A efda4cd163b1
screenshots-memory is a skill published in the GitHub repository mostafa-drz/claude-skills (4 stars, last pushed today), licensed MIT. It adds 212 tokens to every session and 7,138 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-12.
Other skills, from other repositories
shipkit-work-memory
Log session progress and save resume state. Infers from conversation and git. Triggers: 'log progress', 'session summary', 'checkpoint', 'save progress', 'end session'.
shipkit-claude-md
Update CLAUDE.md with learnings, preferences, or project context. Use when user says "update claude.md", "add to claude.md", or wants to persist something.
init-project-memory
Use when the user says "init memory", "set up memory", "scaffold memory", "init project memory", or invokes /init-project-memory — creates the .claude/ session memory system in a new project.
session-update
Use when the user says "update session", "log progress", "checkpoint session", or invokes /session-update — especially before /clear or handoff.
codex-symphony
Install and operate a portable OpenAI Symphony + Linear orchestration setup in any Git repository. Use when the user wants a one-command local Symphony runner, a reusable WORKFLOW template, background launch scripts, or a Codex wrapper that restarts Symphony automatically when reopening the CLI.
napkin
Maintain a per-repo napkin as a continuously curated runbook (not a session log). Activates EVERY session. Read and curate it before work, keep only recurring high-value guidance, organize by priority-sorted categories, and cap each category at top 10 items. The napkin lives at .claude/napkin.md.