goal

goal is a skill for Claude Code from MasonEgger/bpe-claude-code-plugin. It costs 126 tokens per session (3,638 once invoked), scanned A, original, MIT.

An autonomous coding mode that works through a project's planned tasks in `todo.md`. It can run everything, one named section, or one step, while checking branch safety and deciding how the project should be verified.

In plain words
What is it for?
Use it to implement a complete plan, finish one section, or execute a single step with tests or another project verification command.
Why use it?
It removes the need to manually direct the coding agent through every planned task. It also stops early when the branch or working tree is unsafe and can retry failed validation up to three times.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter. Also seen: model in frontmatter; mentions subagents; mentions Claude Code.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the bpe plugin — 13 skills, 3 agents shipped together

Good fit Use it to implement a complete plan, finish one section, or execute a single step with tests or another project verification command.

Compare 6 skills from other repositories ↓
Install

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.

Claude Code
/plugin marketplace add MasonEgger/bpe-claude-code-plugin
Claude Code
/plugin install bpe

Made for: Claude Code.

Or install bpe, the plugin that ships this one along with the rest of its 13 skills, 3 agents.

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.

agentmods badge for goal

README.md
[![agentmods](https://agentmods.dev/badge/skills/masonegger/bpe-claude-code-plugin/goal/github.svg)](https://agentmods.dev/skills/masonegger/bpe-claude-code-plugin/goal)
Your own site
<a href="https://agentmods.dev/skills/masonegger/bpe-claude-code-plugin/goal"><img src="https://agentmods.dev/badge/skills/masonegger/bpe-claude-code-plugin/goal/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.

agentmods 80×15 button for goal

Your own site · 80×15
<a href="https://agentmods.dev/skills/masonegger/bpe-claude-code-plugin/goal"><img src="https://agentmods.dev/badge/skills/masonegger/bpe-claude-code-plugin/goal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 126 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,638 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00126 $0.03638
Opus 5.5 $0.00050 $0.01455
Sonnet 5.5 $0.00025 $0.00728
Haiku 4.5 $0.00013 $0.00364

Measured 1mo ago against content hash f11f3b739eba, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

goal 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 1mo 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.

bpe/skills/goal/SKILL.md · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BPE Autonomous Goal

Set up a /goal-driven autonomous BPE run. The parent session orchestrates; each step is executed by the bpe:step-executor subagent in modes (implement, optional fix loop, finalize). When the step's plan.md section declares validators, bpe:validator runs between implement and finalize and the orchestrator drives the fix loop (max 3 iterations) per ${CLAUDE_PLUGIN_ROOT}/references/validator-protocol.md. Fresh context per dispatch; the parent transcript stays tiny; the /goal evaluator watches it for completion.

Requires Claude Code v2.1.139+. Pairs well with /auto. The /goal command itself must still be run by the user; slash commands cannot invoke other slash commands programmatically, so this command writes the assembled /goal invocation to goal.md at the repo root and you paste it from there.

Mode Routing

Parse $ARGUMENTS:

  • empty or full: autonomous to the end of todo.md. Default. Use when you trust the plan and have time / quota.
  • section <name>: autonomous through one labeled section of todo.md. A section is a top-level Markdown heading or an explicit comment block like <!-- section: foo -->.
  • step: autonomous through ONE step. Rarely the right tool; for a single interactive step, /bpe:execute-plan is simpler. Use step only when you want the autonomous-mode contracts (SHA verification, session-summary-per-commit, push, validator pass) on a single item without driving it yourself.

State the routed mode in user-facing text before pre-flight.

Step 1: Pre-Flight Checks (refuse on any failure)

Run these in parallel and surface results in user-facing text:

git rev-parse --abbrev-ref HEAD
git status --short
test -f plan.md && echo "plan: yes" || echo "plan: NO"
test -f todo.md && echo "todo: yes" || echo "todo: NO"
grep -c '^- \[ \]' todo.md 2>/dev/null || echo 0
grep -q '^commit-msg\.md$' .gitignore && echo "gitignore commit-msg.md: ok" || echo "gitignore commit-msg.md: MISSING"
grep -q '^goal\.md$' .gitignore && echo "gitignore goal.md: ok" || echo "gitignore goal.md: MISSING"
grep -cE '^\*\*(Tools|Validator consults):\*\*' plan.md 2>/dev/null || echo 0

Read the full file on GitHub · 174 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. 1mo ago Changed f11f3b739eba
  2. 2mo ago First seen · 174 lines · 126 tokens per session scan A 18aa89eda7b7

Subscribe to this mod's changes

goal is a skill published in the GitHub repository MasonEgger/bpe-claude-code-plugin (7 stars, last pushed 27d ago), licensed MIT. It adds 126 tokens to every session and 3,638 once invoked, about $0.0005 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 2026-08-31.