What is phoenix-goal?

A workflow that turns a broad coding request into a specific, runnable definition of done before implementation begins. It creates an acceptance check—a test or command that can show whether the requested result was achieved—and a backlog of work.

Use it for vague or high-level goals that need clarification, a measurable outcome, and a path from planning through verified completion.

It prevents building the wrong thing correctly and avoids agents endlessly doing work without a reliable stopping point.

phoenix-goal — Skill for Claude Code and Codex by All-The-Vibes

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 All-The-Vibes/ATV-Phoenix --skill phoenix-goal

Clone the repo

git clone --depth 1 https://github.com/All-The-Vibes/ATV-Phoenix

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/all-the-vibes/atv-phoenix/phoenix-goal/github.svg)](https://agentmods.dev/skills/all-the-vibes/atv-phoenix/phoenix-goal)

Your own site

<a href="https://agentmods.dev/skills/all-the-vibes/atv-phoenix/phoenix-goal"><img src="https://agentmods.dev/badge/skills/all-the-vibes/atv-phoenix/phoenix-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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/all-the-vibes/atv-phoenix/phoenix-goal"><img src="https://agentmods.dev/badge/skills/all-the-vibes/atv-phoenix/phoenix-goal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 183 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,803 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	$0.00183	$0.01803
Opus 5	$0.00092	$0.00901
Sonnet 5	$0.00037	$0.00361
Haiku 4.5	$0.00018	$0.00180

Measured 11d ago against content hash 16e3cdf1912e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

phoenix-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 11d 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/phoenix-goal/SKILL.md · 99 lines

How it starts

The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.

phoenix-goal — from a fuzzy goal to a demonstrated outcome

Goal-oriented autonomous execution (BabyAGI, ReAct, Anthropic's evaluator-optimizer) all share one fatal weakness: without an executable acceptance criterion there is no reliable "done". Naive agents loop forever generating busywork, or hallucinate completion. phoenix-goal fixes this by making the first deliverable an objective done-check, not code.

The most expensive bug is building the wrong thing correctly. The second most expensive is never knowing you're finished. phoenix-goal closes both by formalizing intent into a check that can fail.

The flow

fuzzy goal  ("just go and finish it" — or any goal handed over without a spec yet)
   │
   ▼  FRAME + confirm  ── name the entry, the goal, and the done-check; invite correction before any edit
   │
   ▼  phoenix-think  ── deep interview + research ──▶  Intent Contract
   │                                                  whose DELIVERABLE is one runnable
   │                                                  acceptance check (a real command_exit)
   ▼  write .phoenix-ralph/done-check.json   (MUST start RED — prove it fails today)
   │
   ▼  phoenix-plan   ── decompose ──▶  backlog.json  (each item carries its OWN objective check)
   │
   ▼  hand off to  phoenix-ralph  (the loop)  ──▶  driver proves done-check failure-first ──▶ DONE

The difference from phoenix-ralph: ralph takes a backlog you already have; goal derives the backlog and the acceptance check from a single goal first. Goal is the on-ramp; ralph is the engine.

Steps

FRAME the run — the handshake (always first, before any phoenix-think or edit). The failure this step exists to prevent is silently "operating in the spirit of" a request and wandering off. So open with a short, explicit frame and wait for a go:
- Entry: Phoenix's autonomous skills are phoenix-goal, phoenix-auto (its autopilot), and phoenix-ralph (the persistence loop). If the user arrived via an alias from another harness that maps to none of them, orient in one line — "its autonomous entry is phoenix-goal, so I'll run it that way" — no lecture.
- Goal heard: restate the goal in one sentence, as you understood it.
- Done-check: name the objective acceptance check you're about to FORMALIZE (a real test/build that exercises the goal, which must start RED). This is what "done" will mean.
- Shape + control: one line on how it runs (formalize → decompose → loop; completion proven from the trace) and how to steer or stop it.
- Invite correction before the first edit, then proceed to FORMALIZE.

Read the full file on GitHub · 99 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.

11d ago First seen · 99 lines · 183 tokens per session scan A 16e3cdf1912e

Subscribe to this mod's changes

phoenix-goal is a skill published in the GitHub repository All-The-Vibes/ATV-Phoenix (5 stars, last pushed 7d ago), licensed MIT. It adds 183 tokens to every session and 1,803 once invoked, about $0.0009 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 2026-08-31.

Planning Testing agentic-ai ai-agents developer-tools github-copilot llm-tools mcp

semgrep-rule-variant-creator

Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language.

waybarrios/opencode-power-pack · 50 tokens

build-status-cache

Skip Phase 7 rebuild when no code changed since Phase 5 build passed. Compares git diff hash against stored hash from last successful build validation. Triggers on: entering Phase 7, checking build status, before final build validation. Returns BUILDCACHEHIT to skip or BUILDCACHEMISS to re-run build-validator.

closedloop-ai/claude-plugins · 69 tokens

implementation-standards

Reference material with coding standards (defensive coding, error handling, testing patterns). Loaded on demand by the Developer sub-agent (.github/agents/developer.md); not directly invokable.

attilaszasz/sdd-pilot · 44 tokens

terraform-test

WORKFLOW SKILL — Write and run Terraform tests (.tftest.hcl). WHEN: "create terraform test", "write tftest", ".tftest.hcl", "mock provider", "test module", "test assertion". USE FOR: test files, run blocks, assertions, mock providers, plan-mode unit tests, apply-mode integration tests, test troubleshooting. DO NOT USE…

jonathan-vella/apex-accelerator · 101 tokens

hotfix

Fixes an observed defect with reproducible evidence in one call: writes a short trace doc before touching code, implements the fix, and backs it with a regression test written before the fix. Production incidents are the motivating case, not a gate. When blocked, it halts by name and saves the doc for a later call to…

pe-menezes/vibeflow · 102 tokens

clarification-strategies

Reference material with ambiguity-audit patterns and critique strategies for requirements. Loaded on demand by clarify-spec; not directly invokable.

attilaszasz/sdd-pilot · 33 tokens