project-documentation-structure.rules

project-documentation-structure.rules is a cursor rule for Cursor from drumnation/unsplash-smart-mcp-server. It costs 833 tokens per session, scanned A, original, MIT.

A documentation rule for adding searchable metadata and recording important project knowledge in Markdown files. YAML frontmatter is a small block at the top of a document that labels its title, summary, and keywords.

In plain words
What is it for?
Use it when creating or substantially updating feature, concept, architecture, or other project documentation.
Why use it?
It makes new documentation easier to find and preserves decisions or implementation details that may not be obvious from the code. It also gives documentation a consistent structure.

Cursor rule for Cursor

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 rules/drumnation/unsplash-smart-mcp-server/project-documentation-structure.rules
Clone the repo
git clone --depth 1 https://github.com/drumnation/unsplash-smart-mcp-server

Made for: Cursor.

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 project-documentation-structure.rules

README.md
[![agentmods](https://agentmods.dev/badge/rules/drumnation/unsplash-smart-mcp-server/project-documentation-structure.rules.svg)](https://agentmods.dev/rules/drumnation/unsplash-smart-mcp-server/project-documentation-structure.rules)
Your own site
<a href="https://agentmods.dev/rules/drumnation/unsplash-smart-mcp-server/project-documentation-structure.rules"><img src="https://agentmods.dev/badge/rules/drumnation/unsplash-smart-mcp-server/project-documentation-structure.rules.svg" alt="Measured on agentmods" height="20"></a>
Per session 833 This file is loaded in full into every session.
When invoked 833 The same file — it is already loaded in full.
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 $0.00833 $0.00833
Opus 5 $0.00417 $0.00417
Sonnet 5 $0.00167 $0.00167
Haiku 4.5 $0.00083 $0.00083

Measured 5d ago against content hash 1f54fae916ac, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

project-documentation-structure.rules 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 5d 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.

.cursor/rules/project-documentation-structure.rules.mdc · 43 lines

What it actually says

Documentation Workflow (Agent Action)

  1. Domain Knowledge Capture: When working on a feature or complex task, capture significant domain knowledge (design decisions, tricky implementation details, rationales, API usage notes, error handling strategies, configuration details) that isn't obvious from the code itself.

  2. File Creation & Enhancement for Discoverability:

    • Action: Create new Markdown (.md) files for this knowledge within the appropriate subfolder (e.g., under docs/features/, docs/concepts/, docs/architecture/adr/). Use descriptive kebab-case filenames.
    • Action: Include Structured Metadata (YAML Frontmatter): At the very top of EVERY new or significantly updated documentation file, include or update a YAML frontmatter block. This is critical for search and indexing:
      ---
      title: "[Clear, Descriptive Title of Document]" # Human-readable title
      description: "[One-sentence summary focusing on key concepts and purpose]" # Concise explanation
      keywords: # Comprehensive list for keyword search (grep) and potential RAG indexing
        - "[primary topic]"
        - "[key technology involved, e.g., React, Vitest, Docker]"
        - "[feature name]"
        - "[related component name]"
        - "[core concept]"
        - "[synonym or alternative term]"
        - "[specific function name if relevant]"
        - "[error code if relevant]"
      related_features: ["[feature-folder-name]", "[another-feature]"] # Links to related feature doc folders (use folder names)
      related_concepts: ["[concept-folder-name]"] # Links to related concept doc folders
      related_adr: ["[adr-number]"] # Optional: Link to relevant ADRs
      last_updated: "[YYYY-MM-DD]" # Date of creation/last significant update
      # author: "[Agent/User Name - Optional]"
      ---
      
    • Action: Write a Concise Summary/Abstract: Immediately following the frontmatter, write a short paragraph (2-4 sentences) summarizing the document's main content and purpose. Naturally weave in the most important keywords from the frontmatter. This helps RAG context retrieval.
    • Action: Use Descriptive Headings: Structure the main document content using clear, keyword-rich Markdown headings (##, ###) that accurately reflect the information in each section. This helps both human readers and semantic search.
    • Action: Embed Cross-Links: Where relevant within the main content, add explicit Markdown links ([link text](mdc:relative/path/to/other/doc.md) or [link text](mdc:[root-relative-path])) to other related documentation files within /docs or even specific code files/sections. This strengthens the knowledge graph for RAG.
    • Action: Capture Detailed Knowledge: Populate the rest of the file with the detailed domain knowledge, decisions, implementation notes, etc., captured in Step 1, using clear language and code blocks where appropriate.
  3. Index Updates: CRITICAL: After adding or significantly modifying a documentation file:

    • Action: ALWAYS update the corresponding [foldername].index.md file in its parent directory. Add/update the link using the title from the frontmatter and the description from the frontmatter for context.
    • Action: ALWAYS update the last_updated field in the frontmatter of the modified/created document itself.
    • Action: Consider if the root docs.index.md or higher-level indexes need updates to reflect the new content, especially for major additions. Ensure discoverability through navigation.
  4. Review Existing Docs (Consultation): Before creating new documentation, use the consult-documentation-index rule/process to check relevant index files and potentially perform searches (grep or semantic using keywords) to see if similar information already exists. Update existing documents rather than creating duplicates whenever practical.

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. 5d ago First seen · 43 lines · 833 tokens per session scan A 1f54fae916ac

Subscribe to this mod's changes

project-documentation-structure.rules is a cursor rule published in the GitHub repository drumnation/unsplash-smart-mcp-server (63 stars, last pushed 7mo ago), licensed MIT. It adds 833 tokens to every session, about $0.0042 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 2026-08-30.