agentleFS
SwastikPandey1024SwastikPandey1024/agentic-engineering-os/.agents/skills/documentation-orchestrator/SKILL.md

documentation-orchestrator

Skill1 starsChanged 3 months agoMITView on GitHub

Adaptive project documentation suite generation, classification-based document planning (Utility vs ML vs Fullstack vs Hackathon vs Production), and truth verification.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 10
  2. Distinctive:

    Never generate enterprise-scale documentation for trivial 20-line scripts.

    line 15
  3. Very specific:

    NEVER generate generic filler docs: Avoid creating empty or boilerplate files with TODO: write this section.

    line 99
  4. Distinctive:

    NEVER fabricate benchmarks or test metrics: Use actual recorded values from evaluation runs.

    line 100
  5. Distinctive:

    NEVER leave broken copy-paste commands: Test every installation and run command locally.

    line 101

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Documentation Orchestrator Skill

1. When Should I Use This?

Use this skill when:

  • Determining the appropriate documentation suite and structure for a new or existing repository.
  • Generating complete, synchronized project documentation across business, product, engineering, and operational layers.
  • Auditing documentation to eliminate false claims, aspirational text, or unverified features.

Never generate enterprise-scale documentation for trivial 20-line scripts. Never skip critical architecture and runbook documentation for production applications.


2. What Should I Inspect First?

  1. Project Archetype & Scale:
    • Small CLI / Utility?
    • ML / AI Research or Portfolio project?
    • Full-stack Web Application?
    • Fast-paced Hackathon prototype?
    • Enterprise Production Service?
  2. Current Verified Capabilities: Inspect real endpoints, database tables, models, and test suites to ensure documentation matches code reality 100%.

3. What Workflow Should I Follow?

Classify Project Archetype & Target Audience
                     ↓
Select the Required Minimum Documentation Suite
                     ↓
Audit Code Reality (Verify real functions, routes, configs)
                     ↓
Generate Core Documents Sequentially (High-level → Low-level)
                     ↓
Cross-Link Related Documents & Add Architecture Diagrams
                     ↓
Verify Accuracy: Ensure ZERO unverified or unbuilt features are claimed

The 5 Project Classifications & Minimum Document Suites

┌──────────────────────────────────┬───────────────────────────────────────────────────────────┐
│ Project Classification           │ Required Minimum Documentation Suite                      │
├──────────────────┼───────────────────────────────────────────────────────────┤
│ 1. Small CLI / Script Utility    │ • README.md (Overview, Prerequisites, Setup, Usage)       │
│                                  │ • LICENSE                                                 │
├──────────────────────────────────┼───────────────────────────────────────────────────────────┤
│ 2. ML / AI Portfolio Project     │ • README.md (Hero, Problem, Results, Live Demo)           │
│                                  │ • ARCHITECTURE.md (Data Pipeline, Model Architecture)     │
│                                  │ • EVALUATION_REPORT.md (Metrics, Confusion Matrix, Baselines)│
│                                  │ • MODEL_CARD.md (Intended use, Limitations, Biases)      │
├──────────────────────────────────┼───────────────────────────────────────────────────────────┤
│ 3. Full-Stack Web Application    │ • README.md (Overview, Tech Stack, Local Dev Setup)       │
│                                  │ • PRD.md (Product Requirements & User Stories)            │
│                                  │ • ARCHITECTURE.md (Frontend/Backend/DB Boundaries, Auth)  │
│                                  │ • API.md (Endpoints, Schemas, Status Codes, OpenAPI)      │
│                                  │ • DEPLOYMENT.md (Render/Vercel/Docker Setup, Env Vars)    │
│                                  │ • SECURITY.md (Auth, CORS, Secret Management)             │
├──────────────────────────────────┼───────────────────────────────────────────────────────────┤
│ 4. Hackathon Prototype           │ • README.md (1-Line Hook, Problem, Solution, Demo Video)  │
│                                  │ • PROBLEM_STATEMENT.md & PITCH.md (Judging criteria fit)  │
│                                  │ • ARCHITECTURE.md (High-speed prototype flow)             │
│                                  │ • QUICKSTART.md (1-minute local reproduction guide)       │
├──────────────────────────────────┼───────────────────────────────────────────────────────────┤
│ 5. Enterprise Production Service │ • BRD.md (Business Requirements Document)                 │
│                                  │ • PRD.md (Product Requirements Document)                  │
│                                  │ • SRS.md (Software Requirements Specification)            │
│                                  │ • ARCHITECTURE.md (System design & data flow)             │
│                                  │ • docs/decisions/ (Architecture Decision Records - ADRs)  │
│                                  │ • API.md & OPENAPI schema                                 │
│                                  │ • RUNBOOK.md / DEPLOYMENT.md (Operations, Rollback, CI/CD)│
│                                  │ • SECURITY.md & Threat Model                              │
│                                  │ • OBSERVABILITY.md (Metrics, Traces, SLA/SLO Alerts)      │
└──────────────────────────────────┴───────────────────────────────────────────────────────────┘

4. What Decisions Should I Make?

Documentation DecisionGuiding Principle
Depth vs OverheadRight-size documentation to project longevity. A 2-day hackathon needs high-impact visual presentation; a production service needs rigorous operational runbooks and ADRs.
Truthfulness GateIf a feature is planned for next week, label it in ROADMAP.md, NEVER in README.md or API.md as active.
Visual ArchitectureEvery architecture doc for a non-trivial project must include at least one ASCII or Mermaid sequence/system diagram.

5. What Should I Avoid?

  • NEVER generate generic filler docs: Avoid creating empty or boilerplate files with TODO: write this section.
  • NEVER fabricate benchmarks or test metrics: Use actual recorded values from evaluation runs.
  • NEVER leave broken copy-paste commands: Test every installation and run command locally.

6. How Should I Verify Success?

# 1. Verify all generated documentation files exist and are populated
ls -la docs/ || ls -la README.md ARCHITECTURE.md

# 2. Check for unbroken relative markdown links
python -c "
import os, re
for root, _, files in os.walk('.'):
    for f in files:
        if f.endswith('.md'):
            path = os.path.join(root, f)
            text = open(path, encoding='utf-8').read()
            links = re.findall(r'\[.*?\]\((?!http|mailto)(.*?)\)', text)
            for link in links:
                target = os.path.join(root, link.split('#')[0])
                if not os.path.exists(target) and link.split('#')[0] != '':
                    print(f'Broken link in {path}: {link}')
print('Documentation link validation complete.')
"

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.