danieldekay/ai-ops/dk-flavored-spec-kit

Name: dk-flavored-spec-kit
Author: danieldekay

Trust tier 3

community

Source check pending

Checking that the SKILL.md file is still in the repository.

Shared knowledge base for the dk.speckit.* agent family. Defines evidence-first execution, risk classification, pushback protocol, verification strategy, and artifact contracts. Agents reference this skill to stay DRY — philosophy and formats live here, agent files focus on workflow.

Install

npx vskill@latest install danieldekay/ai-ops/dk-flavored-spec-kit

SKILL.md

dk-Flavored Spec-Kit

Shared knowledge base for the dk.speckit.* agent pipeline. This skill extracts all cross-cutting concerns so individual agent files stay focused on their workflow.

Philosophy

The dk-flavored spec-kit extends the OOTB spec-kit pipeline with three principles borrowed from Anvil:

Evidence-first execution — verification is recorded, not self-reported. If it's not in the ledger, it didn't happen.
Risk-scaled depth — not every task deserves the same scrutiny. Green tasks get a quick check; red tasks get adversarial review.
Pushback as a feature — agents that can say "this is a bad idea" before implementing save more time than agents that implement the wrong thing fast.

Agent Family

Agent	Role	Key Anvil Enhancement
`dk.speckit.main`	Lifecycle orchestrator (GPT-5)	Git hygiene, pushback triage
`dk.speckit.orchestrator`	State detection & routing (Sonnet)	Pushback routing, ledger-aware gates
`dk.speckit.tasks`	Task breakdown with risk labels	🟢🟡🔴 risk + S/M/L sizing
`dk.speckit.implement`	Anvil-enhanced execution	Verification ledger, pushback, reuse detection
`dk.speckit.quality-gate`	Automated checks	Unchanged — reads code
`dk.speckit.autofix`	Bounded fix loops	Ledger-aware context
`dk.speckit.review`	Evidence-backed review	Consumes evidence bundle, adversarial pass for 🔴
`dk.speckit.commit`	Conventional commits	Unchanged
`dk.speckit.retro`	Compound learning	Ledger analytics
`dk.speckit.ui`	UI requirements analysis	Unchanged

How Agents Reference This Skill

Agents load specific reference files before execution:

markdown

## Skill References
 
Before execution, read the following from `skills/dk-flavored-spec-kit/references/`:
-`risk-classification.md` — for interpreting 🟢🟡🔴 labels and S/M/L sizing
-`evidence-ledger.md` — for ledger format and logging rules
-`pushback-protocol.md` — for when/how to escalate via askQuestions

This keeps agent files focused on WHAT to do. The skill defines HOW and WHY.

Pipeline Flow

User Request
    │
    ▼
dk.speckit.main / dk.speckit.orchestrator
    │ (git hygiene → state detection → routing)
    ▼
SPECIFICATION STAGES (OOTB sk-v1, unchanged)
    constitution → specify → clarify → plan → tasks
    │                                         │
    │                          dk.speckit.tasks wraps
    │                          with risk + sizing
    ▼
dk.speckit.implement
    │ (ledger, risk-scaled verify, pushback ↔ orchestrator)
    ▼
dk.speckit.quality-gate ↔ dk.speckit.autofix
    │
    ▼
dk.speckit.review (evidence audit + adversarial for 🔴)
    │
    ▼
dk.speckit.commit → dk.speckit.retro (ledger analytics)

Artifact Flow

tasks.md (with 🟢🟡🔴 labels)
    │
    ├──→ dk.speckit.implement reads risk labels
    │         │
    │         ├──→ evidence-ledger.json (written per task)
    │         └──→ pushback-log.json (if escalations occurred)
    │
    ├──→ dk.speckit.review reads evidence-ledger.json
    │         │
    │         └──→ review-report.md (includes Evidence Audit pass)
    │
    └──→ dk.speckit.retro reads evidence-ledger.json
              │
              └──→ retro.md (includes ledger analytics)

Reference Index

Reference	Purpose	Used By
risk-classification.md	🟢🟡🔴 definitions, S/M/L sizing	tasks, implement, review
evidence-ledger.md	JSON ledger schema, logging rules	implement, review, retro
pushback-protocol.md	When/how to pushback, askQuestions flow	implement, main, orchestrator
verification-strategy.md	Depth scaling by risk level	implement, review
artifact-registry.md	All artifacts, ownership, locations	all agents

Security

Security scanPassed, 90 of 100Pattern checks on the skill's files.
Model reviewPassed

See the full security report

Quality evaluation

Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

SKILL.md in the repositoryCheckingskills/dk-flavored-spec-kit/SKILL.md

Popularity

7-day trend score: 0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

Claude Code
Cursor
GitHub Copilot
Windsurf
Codex
Gemini CLI
Cline
Amp
Roo Code
Goose
Continue
Kilo Code
Junie
Trae
OpenHands
Qwen Code
AI Agent
Tabnine Chat
Sourcegraph Cody
Amazon Q
Firebase Genkit
Sweep
Devin
Phind
Mentat
Claude Desktop
Aider
GPT Pilot
Plandex
bolt.new
v0
Replit Agent
Anse
TypingMind
BetterChatGPT
LibreChat
Warp AI
Pieces
Fig AI

Embed badge

[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/danieldekay/ai-ops/dk-flavored-spec-kit/badge)](https://verified-skill.com/skills/danieldekay/ai-ops/dk-flavored-spec-kit)

Something wrong with this skill or its page? Tell us.