Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add artemnovichkov/skills/plugin install sdlcWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/artemnovichkov/skills/plan)<a href="https://agentmods.dev/commands/artemnovichkov/skills/plan"><img src="https://agentmods.dev/badge/commands/artemnovichkov/skills/plan/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/artemnovichkov/skills/plan"><img src="https://agentmods.dev/badge/commands/artemnovichkov/skills/plan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00023 | $0.00643 |
| Opus 5 | $0.00012 | $0.00321 |
| Sonnet 5 | $0.00005 | $0.00129 |
| Haiku 4.5 | $0.00002 | $0.00064 |
Grade A, and why
plan 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.
How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stage 3 — Nothing implemented without an accepted plan
Design review happens before code generation, not in the PR. The plan must be reviewable on its own: someone who has not read the diff should be able to say "yes, that is the right change" or "no, that breaks X".
$ARGUMENTS is the slug. If empty, list sdlc/spec/ and ask.
Workflow
1. Prefer plan mode
If the session is not in plan mode, tell the user: press shift+tab to enter plan mode, then re-run this command. Plan mode makes the read-only constraint real rather than a promise. If they would rather not, continue in read-only discipline: read and search only, no edits.
2. Load intent and spec
Read sdlc/intent/<slug>.md and sdlc/spec/<slug>.md, plus CLAUDE.md. Then read the actual files the spec names — the plan is only worth something if it names real files and real functions.
Look for existing utilities, helpers, and patterns to reuse. A plan that proposes new code where the repo already has an implementation is a bad plan.
3. Draft the plan
Use ${CLAUDE_PLUGIN_ROOT}/templates/plan.md:
- Files to change, each with what changes and why — real paths, not categories.
- Work order, with the dependency between steps made explicit.
- Tests required: which existing tests cover this, which new ones are needed, what each asserts.
- What could break: neighbouring behaviour, callers, migrations, platform edges.
- Rollback: how this gets undone if it goes wrong.
4. Interrogate it
Before showing it as finished, answer these in the plan itself:
- What breaks if this ships as written?
- What is the riskiest step, and what makes it less risky?
- What is the alternative approach, and why is this one better?
Then hand it to the user and iterate until they accept it.
5. Persist and commit
Write the accepted plan to sdlc/plan/<slug>.md with Status: accepted. If the session used plan mode, copy the approved plan file's content into it rather than rewriting from memory.
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.
- yesterday First seen · 57 lines · 23 tokens per session scan A 17a782093c87
plan is a command published in the GitHub repository artemnovichkov/skills (36 stars, last pushed yesterday), licensed MIT. It adds 23 tokens to every session and 643 once invoked, about $0.0001 per session on Opus 5. 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 2026-09-07.
Other commands, from other repositories
collaborative-spec-builder-existing
Build an aspirational specification for existing code and identify gaps to address.
speckit.tasks
Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.
speckit.taskstoissues
Convert existing tasks into actionable, dependency-ordered GitHub issues for the feature based on available design artifacts.
tracker-sync
Project a flow spec to a tracker (Linear/GitHub/GitLab/Jira) and reconcile body/status/comments two-way — projection, not coordination. NOT plan-sync (/flow-next:sync).
land
Cadence-tick autonomous PR babysitter (CI-fix, resolve, converge, merge, close, release).
create-issue
Create a well-structured GitHub issue with proper metadata.