What is spec-propose?

A planning workflow that turns an agreed software change into OpenSpec documents, including a proposal, design, requirements changes, and implementation tasks. It creates a Git branch, which is a separate line of project history, before writing those documents.

Use it to write a structured proposal for a feature, fix, or refactor before coding begins.

It keeps unimplemented plans off the main branch and makes the intended work, boundaries, and task order explicit. It also validates and commits the planning artifacts.

spec-propose — Skill for Claude Code and Codex by pivoshenko

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.

Any agent

npx skills add pivoshenko/pivoshenko.ai --skill spec-propose

Clone the repo

git clone --depth 1 https://github.com/pivoshenko/pivoshenko.ai

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/pivoshenko/pivoshenko.ai/spec-propose/github.svg)](https://agentmods.dev/skills/pivoshenko/pivoshenko.ai/spec-propose)

Your own site

<a href="https://agentmods.dev/skills/pivoshenko/pivoshenko.ai/spec-propose"><img src="https://agentmods.dev/badge/skills/pivoshenko/pivoshenko.ai/spec-propose/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/pivoshenko/pivoshenko.ai/spec-propose"><img src="https://agentmods.dev/badge/skills/pivoshenko/pivoshenko.ai/spec-propose.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 229 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,821 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00229	$0.02821
Opus 5.5	$0.00092	$0.01128
Sonnet 5.5	$0.00046	$0.00564
Haiku 4.5	$0.00023	$0.00282

Measured 29d ago against content hash 7804a5eef8ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

spec-propose 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 29d ago.

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.

skills/spec-propose/SKILL.md · 142 lines

How it starts

The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Spec Propose

converged idea -> branch -> config rules -> artifacts -> commit.

Planning only. The request that reaches this skill authorizes artifacts, nothing more, even when it is phrased as "build X" or "fix Y". See Planning Boundary.

Order is load-bearing twice: the branch exists before any artifact is written, and the wave rules are installed before tasks.md is written. Both are one-way doors.

Flow

Preflight. openspec context --json. no_openspec_root -> stop, offer openspec init --tools claude, wait for the user. Never auto-init, never fall back to the current directory. Otherwise take root.path as authoritative. User named a store -> openspec store list --json for ids, then --store <id> on every command for the rest of the run
Branch first, via git-branch-create. Type from the change (feat, fix, refactor), desc matching the change name. Why -> artifacts written on the base branch put an unreviewed spec on main, and relocating them afterwards means hand-rewriting the paths in .openspec.yaml on top of the git move
Link the issue, if there is one. The branch carries no ticket id of its own:
- User named an existing issue -> git config branch.<branch>.issue <n>
- No issue, repo has a GitHub remote and uses issues -> offer git-issue-create for this change, then set the key from the number it returns. Offer it, never file one unasked
- No GitHub remote, or the user declines -> skip silently and carry on. Not every repo tracks work in issues, and a spec is perfectly valid without one
Why -> this one git config call is the only thing that makes git-commit's Closes #<n> trailer and git-pr-create's Resolves: line fire later. git-branch-create never writes the key, and nothing downstream can recover the number if it is not set here
Derive the change name. Kebab-case, from the request, and it should read as the branch name without its <type>/ prefix. Name already taken under openspec/changes/ -> ask whether to continue that change (that is spec-apply's job) or start a new one. Never overwrite an existing change directory
Install the wave rules into openspec/config.yaml before the tasks artifact is written. See Wave Rules. This is the step the whole set depends on
Scaffold. openspec new change "<kebab-name>" -> openspec/changes/<name>/.openspec.yaml. The spec-driven schema's artifacts are proposal -> (specs, design) -> tasks, with applyRequires: ["tasks"]
Author the artifacts. See Delegation. Never write artifact content from your own head when openspec instructions can hand you the authoritative guidance - this is a wrapper
Sanity-read tasks.md before committing. See Disjointness Check. Overlap -> fix it now by merging, reassigning ownership, or ordering with needs:
Validate. openspec validate <change> --strict --json -> read items[].issues, fix, re-run until summary.totals is clean
Commit via git-commit: docs(spec): add <change-name> proposal. Artifacts only. Why -> a commit mixing the spec with its implementation makes the spec unreviewable, and spec-apply can no longer revert one without the other
Stop. Print the change name, the branch, and the task count, then hand to spec-apply

Read the full file on GitHub · 142 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

references/tasks-shape.md 421 B

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.

29d ago Changed · -17 lines 7804a5eef8ba
1mo ago First seen · 159 lines · 229 tokens per session scan A 08c3f5832d3c

Subscribe to this mod's changes

spec-propose is a skill published in the GitHub repository pivoshenko/pivoshenko.ai (6 stars, last pushed yesterday), licensed MIT. It adds 229 tokens to every session and 2,821 once invoked, about $0.0009 per session on Opus 5.5. 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 2026-09-16.

Commits Git and workflow OpenSpec Planning agents ai ai-agents ai-skills mcp skills spec

comet-safe-delivery

A Chinese-language procedure for safely delivering specified Comet changes through Git. It covers checking worktrees and unrelated edits, staging exact files, validating hooks, and authorized commits or pushes.

rpamis/comet · 78 tokens

loom-git-workflow

Git operations guidance including branching strategies, commit conventions, merge workflows, conflict resolution, and worktree management.

cosmix/loom · 27 tokens

conventional-commit

Generer conventional commit-meldinger med Nav-relevante scopes og breaking change-format.

navikt/copilot · 22 tokens

commit-changes

Stage, commit, and amend changes with conventional commit messages. Covers reviewing changes, selective staging, writing descriptive commit messages using HEREDOC format, and verifying commit history. Use when saving a logical unit of work to version control, creating a commit with a conventional message, amending the…

pjt222/agent-almanac · 72 tokens

configure-git-repository

Configure a Git repository with proper .gitignore, branch strategy, commit conventions, hooks, and remote setup. Covers initial setup and common patterns for R, Node.js, and Python projects. Use when initializing version control for a new project, adding a .gitignore for a specific language or framework, setting up…

pjt222/agent-almanac · 78 tokens

git-commit

Safely create a git commit by validating repository state, staging intended changes, scanning for secrets/conflicts, generating a Conventional Commits message (repository convention first, else an English Angular default) from the staged diff, and committing without amend.

johnqtcg/awesome-skills · 52 tokens