AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Wf Plan

skill-marchatton-agent-skills-wf-plan · by marchatton

This skill should only be used when the user uses the word workflow and asks to create a commit-ready, deep project plan from a shaped packet (brief, breadboard, risks, spikes) before development starts, with handoff/pickup boundaries to avoid context rot.

No reviews yet
0 installs
32 views
0.0% view→install

Install

$ agentstack add skill-marchatton-agent-skills-wf-plan

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-marchatton-agent-skills-wf-plan)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
6mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Wf Plan? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

wf-plan

Purpose

Turn a shaped packet (wf-shape dossier) into a commit-ready plan without re-litigating the shape.

Never implement production code. Only research and write the plan.

Inputs

Dossier folder path containing:

  • brief.md
  • breadboard-pack.md
  • risk-register.md
  • spike-investigation.md (if present)

Outputs

Inside the same dossier:

  • plan.md

Steps

0) Pickup (recommended if new thread)

  • Invoke pickup if repo/branch state is not fresh.

1) Ingest shaped packet (no idea refinement loop)

  • Read brief → breadboard → risk register → spikes.
  • If perimeter or top risks are missing: route back to wf-shape.

2) Create plan skeleton

  • Create plan.md with:
  • scope (in/out)
  • key flows + key logic
  • acceptance criteria + verification plan
  • sequencing/phases + stop points
  • rollout/rollback + observability
  • dependencies + risks

3) Local research (always)

  • Find similar patterns in repo.
  • Pull institutional learnings (docs/solutions, docs/LEARNINGS.md).
  • Record concrete file paths.

4) External research (conditional)

  • Always external for: security/auth, payments, privacy, external APIs, migrations.
  • Otherwise only if local context is thin.

5) Spec hardening (gap pass)

  • Edge cases, failure modes, concurrency, performance, data integrity, security threats.
  • Update acceptance criteria + verification.

6) Plan review passes

  • Simplicity, risk, ops/release, data integrity, security/privacy, UX/product.
  • Final mandatory pass: invoke oracle on the whole plan and integrate findings (or mark out-of-bounds).

7) Commit gate

  • GO only if:
  • AC measurable
  • verification per AC
  • sequencing explicit
  • rollout/rollback explicit
  • no P1 unknowns remain

8) Handoff to build (recommended boundary)

  • Invoke handoff and include:
  • plan path + summary of phases
  • how to verify
  • rollout notes
  • biggest remaining risks
  • Recommend build in a fresh thread:
  • /new then pickup then run wf-develop or wf-ralph

Verification

  • plan includes acceptance criteria + verification + rollout/rollback
  • oracle pass integrated

Go/No-Go

  • GO if the plan can be built without re-discovering the shape.
  • NO-GO if it depends on “we’ll figure it out during implementation”.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.