technical-specification

A document-writing process that turns finalized product requirements into precise technical contracts. It covers API shapes, data models, state changes, error handling, and testable acceptance criteria.

In plain words
What is it for?
Writing one technical specification per domain in `docs/spec/`, including exact interfaces, data structures, failure responses, and acceptance checks.
Why use it?
It removes ambiguity before implementation, so engineers and coding agents can build against clear rules and boundaries.

Skill for Claude CodeCodex

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 skills/affectionatec/agentic-engineering/technical-specification
Any agent
npx skills add affectionatec/agentic-engineering --skill technical-specification
Clone the repo
git clone --depth 1 https://github.com/affectionatec/agentic-engineering

Made for: Claude Code, Codex.

Per session 96 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,873 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 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 $0.00096 $0.02873
Opus 5 $0.00048 $0.01437
Sonnet 5 $0.00019 $0.00575
Haiku 4.5 $0.00010 $0.00287

Measured yesterday against content hash 6a57fe8cfc7a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

technical-specification scanned grade B with 2 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 yesterday.

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.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

| 2 | **Given** [edge case] **When** [action] **Then** [specific handling] | `curl -s -X POST .../endpoint -d @oversize.json` | response `413`, error code `PAYLOAD_TOO_LARGE` |

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- **Verify via** — the exact command or check the verifier runs (test invocation, curl + assertion, build step)
skills/technical-specification/SKILL.md · 283 lines

How it starts

The opening of the file, as written. The whole thing — 283 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Technical Specification Co-Creation

Core constraint: The spec is the contract. If it's not in the spec, it doesn't get built. If it's ambiguous, ask — don't guess.

Quick Reference

Use when PRD finalized; need precise contracts per bounded context
Skip when PRD still shifting, or for trivial single-file features without architectural weight
Output docs/spec/[domain-name].md — one per domain, never monolithic
Sequence PRD → SPEC → ADR → IMPL PLAN
Iron rule Every number concrete. Every boundary has a violation response. No "etc."
Sibling skills [[project-kickoff-prd]] · [[architecture-decision-record]] · [[implementation-plan]] · [[status-tracker]] · [[independent-verification]] · [[git-workflow]]

Role

You are a senior Systems Architect writing implementation contracts. Your job is to take a finalized PRD and produce specifications so precise that any engineer — or any Agent — can implement them correctly with zero prior context and zero follow-up questions.

Persona & Tone

  • Think like the engineer who will implement this at 2 AM with only this document.
  • Be exhaustively precise — every number, every edge case, every error path.
  • Be structured — consistent format across all spec documents.
  • Challenge hand-waving — if the user says "handle errors gracefully," demand specifics.

When to Create

Create specs after the PRD is finalized (Phase 5 of PRD skill complete). One spec per domain/module. Never write a monolithic spec — split by bounded context.

Conversation Flow

Phase 1: Scope & Decomposition

Based on the PRD, propose how to split specs:

  • Identify bounded contexts / modules from the PRD's feature breakdown
  • Propose one spec document per domain (e.g., spec/auth.md, spec/billing.md, spec/notifications.md)
  • Define the dependency graph between specs (which spec depends on which)

Ask the user to confirm the decomposition before proceeding.

Read the full file on GitHub · 283 lines

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. yesterday First seen · 283 lines · 96 tokens per session scan B 6a57fe8cfc7a

Subscribe to this mod's changes

technical-specification is a skill published in the GitHub repository affectionatec/agentic-engineering (4 stars, last pushed 2mo ago), licensed MIT. It adds 96 tokens to every session and 2,873 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.