claude

spec-agentic

Specification doctrine and toolkit for work executed by coding agents: declared-executor headers, a LOCKED/WORKING/OPEN facts ledger, stable anchor IDs with CI verifiers, an OpenAPI contract set, impact maps, and a read-only spec auditor. Pairs with spec-human.

0 on GitHubby Vasco Maiav0.1.0MITupdated 2 months agoworkflow
/

Plugin spec-agentic from marketplace claude-specs in vmaia00/claude-specs. Commands read from the repo's manifest — 1 month ago.

What it installs

Skills (5)

  • anchor-graph

    Reference across documents by stable named anchors and IDs instead of line numbers, and verify the whole set with a CI gate. Use when cross-references between specs break as documents are edited, when you need "reference by name, not line number", when fighting link rot in a spec set, when building

  • contract-set

    Build an OpenAPI-first contract set as the spine of a spec set — a shared conventions library, one internal contract per downstream capability, and front-facing surface contracts per functional area, each versioned, owned and verifier-gated. Use for API contract design, "define the interface before

  • facts-ledger

    Keep one canonical registry of what is confirmed (LOCKED), what is our own proposal (WORKING) and what is explicitly unconfirmed (OPEN), with a resolvable source citation on every row. Use when tracking confirmed vs assumed facts, answering "what do we actually know", recording client or stakeholder

  • impact-map

    Build per-entity impact cards and touchpoint matrices that answer "what breaks if this changes" for an agent with no working memory, generated from the data model and contract set rather than written by hand. Use when doing impact analysis or blast-radius analysis, when building a traceability or cr

  • spec-for-agents

    Write specifications, PRDs and design documents that a coding agent will implement, using a declared executor contract, exhaustive uniform coverage, per-claim provenance tags, stable grep-resolvable IDs, in-body changelogs and tombstones. Use when authoring a spec that Claude Code or another agent w

Subagents (1)

  • spec-auditor

    Read-only auditor for an agentic spec set. Reports unresolved and orphaned IDs, claims with no provenance tag, fact-tagged claims whose citation does not resolve, contradictions between documents, coverage gaps against the declared document shape, and sections changed without a changelog entry. Retu

View source on GitHub

More from vmaia00/claude-specs

All 7 plugins →

ruby

claude

Curated Ruby/Rails/Hotwire/TDD skills (17), vendored (MIT) from superpowers-ruby + rails-agent-skills + ruby-core-skills, slimmed 2026-08-25. Opt-in; install for Ruby on Rails work.

17 skills

0

Always-on multi-persona code-review pipeline: parallel distinct-angle reviewer panels → a Code Review Master → a second adversarial panel → Master 2 → an opt-in Executor & Fixer. No token optimization; every panelist is always spun up. Opt-in.

1 skill · 1 command · 13 agents

0

core

claude

Orchestrator foundations: generic builder/explorer/planner/reviewer agents, a new-project command, a secret-scan hook, and universal working principles.

5 skills · 4 commands · 4 agents · hooks

0

ecc

claude

Slim remnant of affaan-m/ECC (MIT): 6 curated skills — seo, design-system, api-design, accessibility, make-interfaces-feel-better, postgres-patterns. The other ~580 vendored items were removed 2026-08-25 after a usage audit (re-vendor from upstream if ever needed).

6 skills

0

writing

claude

Document production: latex-doc renders Markdown deliverables into overflow-safe thesis-style PDFs (pandoc + XeLaTeX, three-families overflow method), and pt-pt enforces European Portuguese in every Portuguese response (vendored from mfrade/claude-skills).

2 skills

0

Consulting-grade analysis-framework design: chapter skeleton, named frameworks, per-chapter data requirements, and a visualisation plan from just a research subject; data gathered via ordinary research, report written against the framework with a zero-hallucination policy. Opt-in.

1 skill

0

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Maintain spec-agentic?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins7 plugins