plan-impl

A command that turns a written software specification into an ordered implementation plan. The plan connects requirements to files, tests, risks, and verification tasks.

In plain words
What is it for?
Use it when you have a specification for a feature, bug fix, or other engineering change and need a concrete build-and-test plan.
Why use it?
It makes unfinished requirements and assumptions visible before coding and gives the implementation work a sequence that can be checked.

Command for Cursor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/rhyanvargas/agentic-development-starter-kit/plan-impl
Clone the repo
git clone --depth 1 https://github.com/rhyanvargas/agentic-development-starter-kit

Made for: Cursor.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 461 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00461
Opus 5 $0.00000 $0.00230
Sonnet 5 $0.00000 $0.00092
Haiku 4.5 $0.00000 $0.00046

Measured yesterday against content hash 1714032ca9da, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-impl scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/commands/plan-impl.md · 26 lines

What it actually says

/plan-impl

Create an implementation plan from a specification.

Skill

Read and follow skills/spec-driven-workflow. Resolve paths with references/artifact-homes.md. For Cursor plan files, follow references/cursor-adapter.md (YAML todos required). When the engineering-methods pack is installed, use upstream writing-plans for granular task style — see docs/engineering-methods.md. Before planning Medium+, run or confirm the cross-requirement pass in references/analyze-requirements.md.

Usage

/plan-impl path/to/spec.md

Behavior

  1. Read the spec; if open questions or unconfirmed assumptions remain, resolve those before planning.
  2. Analyze (Medium+): cross-check REQs for conflicts, ambiguity, and gaps (analyze-requirements.md). Update the living spec if findings change REQs. If a plan already exists and REQs changed, treat this as a resync (replace stale todos).
  3. Break work into ordered, verifiable tasks (files, tests, risks). Prefer tasks that map to REQ-XXX and stay within a focused file set. Prefer splitting build from verify/review so QA can run in parallel with the next slice. For bugfix specs, map tasks/tests to both FIX and KEEP (Unchanged) items.
  4. Tracer gate (Large, and Medium when architecture/integration is ambiguous): include a thin vertical slice + one verify before multi-phase implement, or “N/A — architecture proven” with justification — see skill / problem-size-guide.md.
  5. Resolve the plan path via the skill. As a Cursor / command, default to .cursor/plans/{feature-name}.plan.md unless the project already uses docs/plans/ or the user overrides the path.
  6. Write the plan with trackable todos (Cursor: non-empty frontmatter todos; never body-only REQ tables with empty Plan UI todos).
  7. Suggest /implement-spec only after the user approves the plan (medium+).

Keep this command thin — planning depth lives in the skill / recommended upstream.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 26 lines · 0 tokens per session scan A 1714032ca9da

Subscribe to this mod's changes

plan-impl is a command published in the GitHub repository rhyanvargas/agentic-development-starter-kit (2 stars, last pushed 8d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 461 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.