refresh-knowledge

refresh-knowledge is a command for Claude Code from HalidSaglam/saglitzdesign-mcp. It costs 24 tokens per session (464 once invoked), scanned A, original, MIT.

A command for finding outdated project knowledge documents and researching them again before updating them.

In plain words
What is it for?
It helps review documents by age, verify dated claims with current sources, update their content and sources, and record a new update date.
Why use it?
It reduces the risk of relying on old search guidance, design standards, marketing practices, or interface patterns.

Command for Claude Code

Part of the saglitzdesign plugin — 7 skills, 9 commands, 2 MCP servers shipped together

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 commands/halidsaglam/saglitzdesign-mcp/refresh-knowledge
Clone the repo
git clone --depth 1 https://github.com/HalidSaglam/saglitzdesign-mcp

Made for: Claude Code.

Or install saglitzdesign, the plugin that ships this one along with the rest of its 7 skills, 9 commands, 2 MCP servers.

Wrote 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.

agentmods badge for refresh-knowledge

README.md
[![agentmods](https://agentmods.dev/badge/commands/halidsaglam/saglitzdesign-mcp/refresh-knowledge.svg)](https://agentmods.dev/commands/halidsaglam/saglitzdesign-mcp/refresh-knowledge)
Your own site
<a href="https://agentmods.dev/commands/halidsaglam/saglitzdesign-mcp/refresh-knowledge"><img src="https://agentmods.dev/badge/commands/halidsaglam/saglitzdesign-mcp/refresh-knowledge.svg" alt="Measured on agentmods" height="20"></a>
Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 464 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found 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 $0.00024 $0.00464
Opus 5 $0.00012 $0.00232
Sonnet 5 $0.00005 $0.00093
Haiku 4.5 $0.00002 $0.00046

Measured 3d ago against content hash 8c094d2355a0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

refresh-knowledge 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 3d 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.

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.

.claude/commands/refresh-knowledge.md · 22 lines

What it actually says

Refresh the SaglitzDesign knowledge base. Follow this procedure:

  1. Find stale docs. Compute each doc's age from its updated: frontmatter field (all docs live under knowledge/**/*.md). Staleness thresholds by category: seo/geo 120 days; design-language/marketing 240 days; pattern 300 days; component/ux/craft/process 365 days; book 730 days. (Same thresholds as the server's knowledge_freshness tool — you can also run the built server and call that tool.) If the user passed arguments (e.g. a category or doc id), refresh only those instead.

  2. Re-research each stale doc with parallel agents. For each stale doc, launch a general-purpose agent that:

    • Reads the current doc.
    • Uses WebSearch/WebFetch to verify every dated claim (thresholds, specs, version-specific guidance) against current official sources; for pattern docs, re-runs the original Mobbin MCP searches (load tools via ToolSearch "select:mcp__mobbin__search_screens,mcp__mobbin__search_flows,mcp__mobbin__search_sections").
    • Edits the doc in place: corrects outdated facts, adds significant new developments, keeps the existing frontmatter format, updates the sources array, and sets updated: to today's date.
    • Reports back a one-paragraph changelog (what changed, what was verified-unchanged).
  3. Refresh the visual example library if stale (>300 days): re-run the Mobbin collection prompts (see git history for the original agent prompts), replacing dead images in knowledge/examples/.

  4. Verify. Run npm run build and start the server once to confirm all docs still parse (stderr reports the loaded doc count — it must not drop). Spot-check one updated doc via a get_design_doc JSON-RPC call.

  5. Commit with a message summarizing the changelog per doc.

Do not rewrite docs wholesale — surgical updates only; the curated structure and prescriptive tone must survive.

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. 3d ago First seen · 22 lines · 24 tokens per session scan A 8c094d2355a0

Subscribe to this mod's changes

refresh-knowledge is a command published in the GitHub repository HalidSaglam/saglitzdesign-mcp (3 stars, last pushed 5d ago), licensed MIT. It adds 24 tokens to every session and 464 once invoked, about $0.0001 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 2026-08-31.