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

Plan Save

skill-samuelbostic29-claude-skills-plan-save · by SamuelBostic29

|

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

Install

$ agentstack add skill-samuelbostic29-claude-skills-plan-save

✓ 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-samuelbostic29-claude-skills-plan-save)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo 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 Plan Save? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Plan Save: Structure and persist a plan for iterative execution

You have just helped create a plan (either through plan mode or conversation). Your job is to restructure it into a persistent, trackable format and save it to a file.

When to use this skill

  • Plan mode (or a planning conversation) just produced a plan, and the work will span multiple sessions.
  • "Save this plan" / "write the plan to a file" / "make this plan trackable".

When NOT to use this skill

  • No plan exists in the conversation yet — there is nothing to restructure; plan first.
  • The plan is already saved — edit the file directly, or use plan-next (execute) / plan-review (critique).
  • A small, single-session change — just do the work; a plan file adds ceremony, not value.

Steps

  1. Identify the plan in the current conversation. Look for the plan that was just discussed or created in plan mode.
  1. Ask the user for the save location using AskUserQuestion. Suggest a descriptive filename in the current project directory (e.g., ./dev-mode-plan.md, ./auth-refactor-plan.md). Do NOT save to ~/.claude/.
  1. Restructure the plan into the format described below.
  1. Write the file to the location the user specified.

Output format

# [Plan Title]

## Context

[Extract the background, goals, constraints, architecture decisions, and any other
relevant context from the plan. This section should give a future session enough
information to understand WHY this plan exists and WHAT is being built. Include
file paths, architecture diagrams, design principles — anything a fresh session
needs to pick up the work without re-exploring the codebase.]

## Implementation phases

Phases are executed one at a time across conversation sessions. When a phase is finished,
add an `#### Implemented` subsection summarizing what was done, then prefix the phase title
with **COMPLETED** (e.g., `### COMPLETED 1. Setup database schema`).
On a new session, read this file first — skip any COMPLETED phases and pick up the next open one.
Do not automatically move to the next phase when done — mark the phase completed and stop.

### 1. [Phase title]

[What needs to be done — files to create/modify, specific implementation details,
code snippets, acceptance criteria. Be detailed enough that a fresh session can
execute this without guessing.]

### 2. [Phase title]

[...]

### N. [Phase title]

[...]

Rules

  • Preserve ALL technical detail from the original plan — do not summarize away specifics like file paths, code snippets, architecture notes, or design rationale.
  • The Context section should be comprehensive enough for a completely fresh session (no prior conversation history) to understand the project and make good decisions.
  • Each phase should be a coherent, self-contained unit of work that can be completed in one session.
  • If the original plan already has numbered sections, steps, or phases, use those as the phases. Preserve the original numbering/naming style where it makes sense.
  • If the original plan doesn't have clear steps, organize it into logical phases yourself based on dependency order.
  • Keep phase descriptions detailed — include file paths, code snippets, architecture notes, expected behavior.
  • Do NOT add phases that weren't in the original plan.
  • Do NOT remove or reword technical content to make it "cleaner" — accuracy matters more than aesthetics.

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.