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.
npx skills add hoatv2211/GameStudio-CodexKIT --skill studio-goal-progressgit clone --depth 1 https://github.com/hoatv2211/GameStudio-CodexKITWrote 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/skills/hoatv2211/gamestudio-codexkit/studio-goal-progress)<a href="https://agentmods.dev/skills/hoatv2211/gamestudio-codexkit/studio-goal-progress"><img src="https://agentmods.dev/badge/skills/hoatv2211/gamestudio-codexkit/studio-goal-progress.svg" alt="Measured on agentmods" 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 | $0.00036 | $0.01650 |
| Opus 5 | $0.00018 | $0.00825 |
| Sonnet 5 | $0.00007 | $0.00330 |
| Haiku 4.5 | $0.00004 | $0.00165 |
Grade A, and why
studio-goal-progress 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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Studio Goal Progress
Overview
Operate an explicitly opted-in KIT Goal from one portable append-only event stream. Percentage is verified weighted work; ETA is a separate range and confidence forecast. Compact and full GUI views read the same reduced state and never control the Goal.
When to use
Use to preflight or initialize an explicit KIT-managed Goal, append a meaningful workflow event, recover current state from progress.jsonl, report verified percentage and ETA, or handle an operator request to Open Progress.
When NOT to use
Do not use for arbitrary Codex tasks, packet planning, speculative progress, durable session transfer, or automatic dashboard opening. Route packet construction to feature-to-work-packets and durable transfer to studio-handoff.
Required inputs and context discovery
Require an explicit Goal opt-in, one goal ID, repository root and snapshot, plan version, scope and do-not-touch paths, ordered packets, final verification requirements, and max_active_timed_packets: 1. Runtime bindings are optional; declared bindings must identify a Codex thread or Hermes session. Every packet requires workflow_id, owner, positive integer progress_weight, dependencies, completion criteria, evidence kinds, and may omit or null its low/high duration estimate.
Safety and risk level
The helpers write only beneath the selected ignored goal root. Run preflight before init. Because init may launch a detached private writer runtime, obtain explicit service-start approval immediately before invoking it; Goal opt-in and a passing preflight are not approval. One writer appends validated events; workflows submit candidates instead of editing JSONL. GUI, context, and handoff consumers are read-only. Never expose raw logs, unrestricted paths, credentials, runtime keys, reusable session tokens, or mutation controls. Browser or panel opening separately requires an explicit Open Progress request.
Workflow
- Confirm explicit KIT Goal participation and run report-only
preflightagainst the manifest and bounded goal root before initialization. Completion criterion: schema, root, snapshot, packet ownership, positive weights, dependencies, required evidence, any declared runtime bindings, and timing policy are valid; otherwise initialization isBLOCKED. - Immediately before
init, obtain explicit approval to start or reuse the private writer service, then initialize one goal-local writer from the accepted preflight inputs. Do not open a GUI. Completion criterion: without immediate service-control approval initialization remainsBLOCKED; with approval,goal.startedis accepted intoprogress.jsonl, reduced state is reproducible, and the Goal remains separate from other local sessions. - Submit candidate events only at meaningful boundaries: Goal start, packet transition, evidence observation, plan revision, context refresh, terminal state, or an explicitly bounded heartbeat for a long operation. Completion criterion: ordinary commentary, elapsed wall time, raw tool output, and optional runtime observations never become completion evidence.
- Reduce the full append-only stream and calculate percentage from verified weight divided by active planned weight. Recalculate on approved plan revision without changing retained verified weight.
Completion criterion: packet verification cites accepted matching
Verifiedevidence, incomplete Goals display below 100%, and replay restores current state. - Calculate ETA as a low/high seconds range with basis, calibration, confidence, and warnings. Missing estimates return
Calculating...;waiting_input,blocked, or stale state pauses ETA; unsupported parallel timing isBLOCKED. Completion criterion: ETA remains forecast rather than evidence and no inactive time or commentary implies completion. - On
Open Progress, accept an explicit goal root, or first resolve one unique nonterminal Goal from the caller's declared manifest runtime binding before repository-level ambiguity fallback. Reuse a healthy authenticated dashboard runtime without new service approval. Starting a stopped or missing runtime requires immediate explicit service-control approval. Python returns one one-time localhost URL and separate server, panel, and browser statuses; it never invokes UI or browser APIs. After URL generation, the host agent uses its panel tool when available, otherwise instructs the operator to open the URL manually. Completion criterion: one top-level one-time URL, host-panel instruction, manual-browser instruction, and independent integration statuses are returned without reusable credentials; ambiguous selection, missing service approval, failed server start, unavailable panel/browser, or unavailable App Server is labeledBLOCKEDwhile portable progress remains usable. - At terminal completion, require every active packet and every final verification requirement to have accepted matching evidence, then return a final evidence card. Completion criterion: the card records goal ID, plan version, verified percentage, ETA status, event and reduced-state artifacts, commands and exit codes, integration limitations, evidence labels, final verdict, and next action.
What ships with it
13 files 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.
- agents/openai.yaml 313 B
- assets/app.js 13 KB runs code
- assets/index.html 6.4 KB
- assets/styles.css 9.0 KB
- references/commands.md 5.5 KB
- schemas/studio-goal-event.schema.json 26 KB
- schemas/studio-goal-manifest.schema.json 5.2 KB
- schemas/studio-goal-state.schema.json 13 KB
- scripts/goal_progress_app_server.py 5.6 KB runs code
- scripts/goal_progress_core.py 52 KB runs code
- scripts/goal_progress_server.py 111 KB runs code
- scripts/goal_progress_store.py 26 KB runs code
- scripts/goal_progress.py 22 KB runs code
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 · 87 lines · 36 tokens per session scan A 5d3eb3bcf15e
studio-goal-progress is a skill published in the GitHub repository hoatv2211/GameStudio-CodexKIT (3 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 1,650 once invoked, about $0.0002 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 2026-09-06.
Other skills, from other repositories
codex-sdd:plan
Create design.md with technical architecture and tasks.md with executable breakdown from an approved proposal. Second phase of OpenSpec workflow.
codex-sdd:tapd-openspec-proposal
Fetch TAPD story requirements before OpenSpec proposal generation. Use when the user asks OpenSpec propose with a TAPD story URL/id, says to import TAPD requirements, or asks to generate proposal.md from TAPD.
agent-swarm-issue
Agent skill for swarm-issue - invoke with $agent-swarm-issue.
agent-issue-tracker
Agent skill for issue-tracker - invoke with $agent-issue-tracker.
agent-project-board-sync
Agent skill for project-board-sync - invoke with $agent-project-board-sync.
agent-user-tools
Agent skill for user-tools - invoke with $agent-user-tools.