agentleFS
Abdox-menaraAbdox-menara/guard-skills/skills/workflow/technical-specification/SKILL.md

technical-specification

Skill0 starsChanged 3 months agoMITView on GitHub

ULTRA-ADVANCED Technical Specification - Technical specification - architecture decisions, API contracts, data model, security review CAPABILITIES: - 8 anti-pattern detection patterns with severity scoring - Compliance scoring (A-F grade) with weighted severity model - Standards-based detection (Google Design Docs; ADR; RFC 2119) - Automated fix suggestion generation - Directory scanning with aggregated reporting TRIGGER PHRASES: "technical specification, tech spec, design document" STANDARDS: Google Design Docs; ADR; RFC 2119 ENVIRONMENT: Works with any codebase, any language, any framework. SECURITY: Read-only analysis.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Technical Specification - ULTRA-ADVANCED v2.0

Overview

Technical specification - architecture decisions, API contracts, data model, security review. This guard scans codebases for violations, scores compliance, and generates actionable remediation recommendations.

Detection Patterns

PatternDescriptionSeverity
Missing spec for complex feature>5 days without spechi
API contracts not definedNo request/response schemahi
Data model undocumentedSchema changes undescribedhi
Security not addressedSecurity impact ignoredcr
Migration plan missingNot specifiedhi
No alternativesSingle solutionmd
Open questions untrackedUnresolved itemsmd
No reviewNot reviewed before implhi

Standards

  • Google Design Docs; ADR; RFC 2119

Scoring

GradeScoreMeaning
A90-100Excellent
B75-89Good
C60-74Fair
D40-59Poor
F0-39Failing

Implementation

import re, os
from typing import Dict, List, Set, Any
from datetime import datetime

class TechnicalSpecification:
    def __init__(self):
        self.file_results = {}
    def scan_file(self, fp: str) -> Dict:
        try:
            with open(fp, "r", encoding="utf-8", errors="ignore") as f:
                c = f.read()
            return self._analyze(c, fp)
        except Exception as e:
            return {"error": str(e), "file": fp}
    def _analyze(self, c: str, fp: str) -> Dict:
        issues = []
        for pid, info in self._patterns().items():
            m = re.findall(info["rx"], c, re.IGNORECASE | re.MULTILINE)
            if m:
                issues.append({"id": pid, "sev": info["sev"], "cnt": len(m), "msg": info["msg"]})
        s = max(0.0, 100.0 - sum(self._w(i["sev"]) * i["cnt"] for i in issues))
        return {"file": fp, "issues": issues, "score": round(s, 1), "grade": self._g(s), "at": datetime.now().isoformat()}
    def _patterns(self):
        return {}
    def _w(self, s):
        return {"cr": 10.0, "hi": 7.0, "md": 5.0, "lo": 2.0}.get(s, 5.0)
    def _g(self, s):
        return "A" if s >= 90 else "B" if s >= 75 else "C" if s >= 60 else "D" if s >= 40 else "F"
    def scan_dir(self, d, exts=None):
        r = {}
        sk = {".git", "node_modules", "venv", "__pycache__", "target", "build", "dist"}
        for root, dirs, files in os.walk(d):
            dirs[:] = [dd for dd in dirs if not dd.startswith(".") and dd not in sk]
            for f in files:
                if exts and not any(f.endswith(e) for e in exts): continue
                r[os.path.join(root, f)] = self.scan_file(os.path.join(root, f))
        return r
    def report(self, r):
        if not r: return {"error": "No results"}
        ti = sum(len(x.get("issues", [])) for x in r.values() if "issues" in x)
        sc = [x.get("score", 0) for x in r.values() if "score" in x]
        a = sum(sc) / len(sc) if sc else 0
        return {"files": len(r), "issues": ti, "avg": round(a, 1), "grade": self._g(a),
                 "cr": sum(1 for x in r.values() for i in x.get("issues", []) if i.get("sev") == "cr"),
                 "hi": sum(1 for x in r.values() for i in x.get("issues", []) if i.get("sev") == "hi"),
                 "recs": []}

Version: 2.0.0 Status: PRODUCTION READY Patterns: 8 Standards: Google Design Docs; ADR; RFC 2119

See Also

  • specification-by-example
  • technical-debt

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.