Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

3 Plan

ASecurity

Phase 3 — Task breakdown, materials list, and budget. Outputs: home_plans/<project>/plan.md

8 stars
0 votes
0 copies
0 views
Added 9/20/2026
ai-agentsgo

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add tstapler/dotfiles --skill 3-plan --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of 3 Plan?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for 3 Plan
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tstapler-3-plan/badge)](https://www.skillsdirectory.com/skills/tstapler-3-plan)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
description: "Phase 3 — Task breakdown, materials list, and budget. Outputs: home_plans/<project>/plan.md"
user-invocable: true
---

# home:3-plan

Dispatch a planning subagent to produce the full project plan — task sequence, materials list, budget, and timeline. Then run an adversarial reviewer to catch gaps before work begins.

## Instructions

1. **Follow [SETUP.md](SETUP.md)** — identify PROJECT_NAME, and mark this stage's task `in_progress` if a task tool is available (Check 3).

2. **Entry gate — read all inputs** (coordinator reads these to pass to subagent):
   - `home_plans/<PROJECT_NAME>/scope.md` — halt if missing, run `/home:1-scope` first
   - `home_plans/<PROJECT_NAME>/research/methods.md` — warn if missing, continue with scope only
   - `home_plans/<PROJECT_NAME>/research/materials.md`
   - `home_plans/<PROJECT_NAME>/research/logistics.md`
   - `home_plans/<PROJECT_NAME>/research/risks.md`

3. **Dispatch a planning subagent using the `Agent` tool.**

   The subagent prompt must include:
   - Full text of `scope.md`
   - Full text of all research files (if present)
   - These exact instructions:

   > You are a home project planning subagent. Produce a complete, actionable project plan grounded in the research provided — do not invent facts the research doesn't support.
   >
   > **Step 1:** Review the scope and research. Identify whether any permits are required — if so, note them prominently at the top of the plan.
   >
   > **Step 2:** Write `home_plans/<PROJECT_NAME>/plan.md` following the template below. Use concrete product names and quantities — no vague references. Tasks should be sized for a single work session (30 min–3 hrs each). Where a research file linked a product to a wiki page (`[[Product Name]]`), reuse that same wiki-link in the Materials List rather than plain text — this keeps the plan connected to the wiki's product/retailer graph. Carry forward source citations from the research files: every non-obvious budget, code, or safety claim in the plan should be traceable to a source in the research, listed in the plan's trailing `## Sources` section.
   >
   > **Step 3:** Return a summary: phase count, task count, total budget estimate, whether permits are required.

   Plan template:
   ```markdown
   # Project Plan: <PROJECT_NAME>

   **Date**: <YYYY-MM-DD>
   **Project**: <one-line description>
   **Status**: Ready to execute
   **Permits Required**: Yes/No — <type and estimated cost if yes>
   **Estimated Total Cost**: $<low>–$<high>
   **Estimated Duration**: <X days / X weekends>

   ---

   ## ⚠️ Safety Precautions
   <Non-negotiable safety steps — utilities to shut off, PPE required, hazards to avoid>

   ---

   ## Task Sequence

   ### Phase 1: Prep & Procurement
   #### Task 1.1: <name> (~<duration>)
   - <Step-by-step actions>
   - **Done when**: <specific observable completion criterion>

   #### Task 1.2: <name> (~<duration>)
   ...

   ### Phase 2: <Core Work Phase Name>
   #### Task 2.1: <name> (~<duration>)
   ...

   ### Phase 3: Finishing & Cleanup
   #### Task 3.1: <name> (~<duration>)
   ...

   ---

   ## Materials List

   | Item | Spec / Model | Qty | Unit Cost | Total | Source |
   |------|-------------|-----|-----------|-------|--------|
   | <item> | <specific product or spec> | <N> | $<X> | $<X> | Lowe's / Amazon / etc. |

   **Materials subtotal**: $<X>

   ---

   ## Tools Required

   | Tool | Own / Rent / Buy |
   |------|-----------------|
   | <tool> | Own |
   | <tool> | Rent from Home Depot (~$<X>/day) |

   ---

   ## Budget Breakdown

   | Category | Estimate |
   |----------|---------|
   | Materials | $<X> |
   | Tool rental | $<X> |
   | Permits | $<X> |
   | Contractor labor (if any) | $<X> |
   | Contingency (15%) | $<X> |
   | **Total** | **$<X>–$<X>** |

   ---

   ## Timeline

   | Day / Weekend | Work |
   |--------------|------|
   | Day 1 | <tasks> |
   | Day 2 | <tasks> |

   ---

   ## Contractor Requirements
   <If any work requires a licensed contractor: what license, what scope, and what to verify before hiring>
   <"None required" if fully DIY>

   ---

   ## Decisions & Open Questions
   <Any choices that need to be made before or during execution>

   ---

   ## Sources
   <URLs carried forward from the research files, backing the budget/code/safety claims above>
   ```

4. **Wait for the subagent to complete.** Do not continue until plan.md has been written.

5. **Dispatch an adversarial reviewer subagent using the `Agent` tool.**

   The subagent prompt must include:
   - Full text of `plan.md`
   - Full text of `scope.md`
   - These exact instructions:

   > You are an adversarial reviewer for a home project plan. Find gaps before any work begins.
   >
   > Review for:
   > 1. **Missing safety steps** — Are there hazards (electrical, plumbing, structural, chemical) without explicit precautions? Should utilities be shut off? Is PPE specified?
   > 2. **Permit gaps** — Does the scope require a permit that isn't noted? (Electrical work, structural changes, plumbing changes, HVAC modifications often require permits.)
   > 3. **Budget underestimates** — Are costs realistic? Are common hidden costs (demo surprises, disposal fees, primer coats, drying time, re-work) accounted for? Is the contingency sufficient?
   > 4. **Sequencing errors** — Are any tasks ordered incorrectly? (e.g., painting before priming, installing before measuring, ordering materials without lead time buffer)
   > 5. **Missing materials** — Are consumables (sandpaper, tape, drop cloths, fasteners, caulk, primer) accounted for?
   > 6. **Scope creep** — Are any tasks broader than what the scope asked for?
   > 7. **Pet and occupant safety** — Are there steps requiring the space to be vacated? Toxic fumes? Sharp debris?
   > 8. **Missing citations** — Are there budget, code, or safety claims stated as fact with no source in the plan's `## Sources` section or the underlying research?
   >
   > For each concern, classify as:
   > - **BLOCKER** — Must be resolved before work starts (safety hazard, required permit, impossible task sequence)
   > - **CONCERN** — Should be addressed; will likely cause problems if ignored
   > - **MINOR** — Low impact; note it but don't block
   >
   > Write findings to `home_plans/<PROJECT_NAME>/adversarial-review.md`:
   >
   > ```markdown
   > # Adversarial Review: <PROJECT_NAME>
   >
   > **Date**: <YYYY-MM-DD>
   > **Verdict**: BLOCKED / CONCERNS / CLEAN
   >
   > ## Blockers
   > - [ ] <issue> — <recommendation>
   >
   > ## Concerns
   > - [ ] <issue> — <recommendation>
   >
   > ## Minors
   > - <issue>
   > ```
   >
   > Return a one-line summary: verdict + count of blockers/concerns/minors.

6. **Wait for the adversarial reviewer to complete.** Read the summary.

   - **BLOCKED** → read adversarial-review.md, patch plan.md to resolve each BLOCKER, then re-run the adversarial reviewer on the updated plan (repeat until CONCERNS or CLEAN).
   - **CONCERNS or CLEAN** → proceed.

7. **Exit gate**: confirm `plan.md` and `adversarial-review.md` both exist and are non-trivial, and that the review verdict is CONCERNS or CLEAN (never advance past a BLOCKED verdict). Mark this stage's task `completed` if a task tool is available, then output the coordinator summary:
   ```
   ✅ Phase 3 complete — plan.md written to home_plans/<PROJECT_NAME>/

   Phases: <N> | Tasks: <N>
   Budget estimate: $<X>–$<X>
   Permits required: Yes/No
   Adversarial review: <BLOCKED|CONCERNS|CLEAN> — <N> blockers, <N> concerns, <N> minors

   Next step: /home:4-prep
   ```

Attribution

tstaplertstapler
View sourceMore from tstapler →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →