speckit.commit

A command that creates a local commit using the Conventional Commits style, a shared format such as feat or fix followed by a short description. It derives the message from the matching feature plan and never pushes to the remote repository.

In plain words
What is it for?
Use it to check the branch and working tree, read specs/plan.md, infer the commit type, and create a local commit from staged or unstaged changes.
Why use it?
It avoids manually composing commit messages and prevents committing when there are no changes or no valid feature branch.

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/visactor/vchart/speckit.commit
Clone the repo
git clone --depth 1 https://github.com/VisActor/VChart

Made for: Cursor.

Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,252 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 $0.00026 $0.01252
Opus 5 $0.00013 $0.00626
Sonnet 5 $0.00005 $0.00250
Haiku 4.5 $0.00003 $0.00125

Measured 2d ago against content hash 5bcd4e557e23, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

speckit.commit 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 2d 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.

.cursor/commands/speckit.commit.md · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

User Input

$ARGUMENTS

Outline

  1. Pre-flight checks (local-only):

    • Treat this command as zero-input by default: if $ARGUMENTS is empty, it should still be able to run end-to-end.
    • Verify the current branch name follows the feature convention (e.g. 001-some-feature) so it can be mapped to a specs/<spec-id>/ directory.
    • Confirm there are staged or unstaged changes in the working tree; if there are no changes, abort without creating a commit.
    • Interpret any CLI flags only as behaviour toggles (e.g. --edit, --no-verify) and not as a manual commit message.
  2. Load spec context from specs/:

    • Derive the feature prefix from the current branch name and resolve the matching specs/<spec-id>/ directory using prefix-based lookup (e.g. 001-specs/001-...).
    • Read specs/<spec-id>/plan.md and extract:
      • The ## Summary section — first non-empty line for the natural-language summary.
      • The top-level title (e.g. Implementation Plan: ...) as a fallback description.
      • The ### Source Code (repository root) code block, which lists the source layout for this feature.
  3. Infer type from Summary keywords:

    • Build a short text corpus from the Summary sentence and plan title.
    • Using only that text, map the commit type according to the following rules (case-insensitive):
      • Contains bug or fixfix
      • Contains refactor or refactoringrefactor
      • Contains perf or performanceperf
      • Contains docs, doc, or documentationdocs
      • Contains test, tests, or unittest
      • Contains implement, implementation, add, support, or featurefeat
      • Otherwise → chore
    • Do not inspect file paths, git history, or common/changes when choosing the type.
  4. Infer scope from Source Code paths:

    • From the Source Code (repository root) block in plan.md, parse all paths that live under packages/<name>/....
    • Collect the second-level directory names (e.g. packages/vchart/vchart, packages/vchart-types/vchart-types), de-duplicate, and sort.
    • Build the scope as:
      • A comma-separated list of all such package names when one or more packages are present.
      • The literal string monorepo when no packages/* paths can be found.
    • Do not attempt to infer scope from other folders or git diffs.

Read the full file on GitHub · 79 lines

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. 2d ago First seen · 79 lines · 26 tokens per session scan A 5bcd4e557e23

Subscribe to this mod's changes

speckit.commit is a command published in the GitHub repository VisActor/VChart (1,834 stars, last pushed 7d ago), licensed MIT. It adds 26 tokens to every session and 1,252 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 2026-08-30.