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 cskwork/oh-my-symphony --skill supergoalgit clone --depth 1 https://github.com/cskwork/oh-my-symphonyWrote 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/cskwork/oh-my-symphony/supergoal)<a href="https://agentmods.dev/skills/cskwork/oh-my-symphony/supergoal"><img src="https://agentmods.dev/badge/skills/cskwork/oh-my-symphony/supergoal/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.
<a href="https://agentmods.dev/skills/cskwork/oh-my-symphony/supergoal"><img src="https://agentmods.dev/badge/skills/cskwork/oh-my-symphony/supergoal.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00024 | $0.00739 |
| Opus 5.5 | $0.00010 | $0.00296 |
| Sonnet 5.5 | $0.00005 | $0.00148 |
| Haiku 4.5 | $0.00002 | $0.00074 |
Grade A, and why
supergoal 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.
How it starts
The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Supergoal factory runtime
One objective -> smallest correct change -> verified against ground truth. This pinned router intentionally exposes only the routes emitted by Oh My Symphony's autonomous factory; every advertised support file ships with the factory.
Standing rules
- Read
.supergoal/rules/RULES.mdfirst when it exists. Usereference/rules.mdfor the standing-rules contract. - Ground truth beats proxy. Run the real proof commands named by the ticket.
- Keep changes surgical, test first, and report blocked proof honestly.
- Destructive or outward-facing actions require explicit operator consent.
Route
| Ticket signal | Mode | Runtime |
|---|---|---|
| map a product, decompose work, identify the next frontier | WAYFINDER | reference/wayfinder.md; no product code |
| build a new vertical slice | GREENFIELD | reference/role-loop.md |
| fix broken behavior | DEBUG | reference/debugging.md, then reference/role-loop.md |
| add or change an existing system | LEGACY | reference/domain-context.md, then reference/role-loop.md |
WAYFINDER may use reference/research.md when a ticket needs external evidence.
Delivery work uses reference/delivery-gate.md and reference/qa.md. When
persisted data is load-bearing, use reference/db-access.md. Optional board
observability is defined by reference/observability.md.
Delivery loop
- Frame the approved ticket in
GOAL.md,PLAN.md,QA.md, andrun-state.jsonfromtemplates/. - Build in a fresh isolated worktree. Bugs start with a failing reproduction.
- Improve the full ticket, then edge cases.
- Run a no-source-edit adversarial review.
- Re-run exact proof. Unmet criteria go to
R-LOOP.md; complete runs writeZ-DONE.md. Never commit beforetemplates/commit-gate.shis green.
Read reference/role-loop.md for the complete execution and isolation
contract. Runtime roles live in agents/executor.md,
agents/code-reviewer.md, agents/qa-auditor.md, agents/qa-tester.md, and
agents/security-reviewer.md. DEBUG also uses agents/debugger.md; LEGACY
discovery uses agents/explore.md.
What ships with it
60 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/code-reviewer.md 2.7 KB
- agents/db-reader.md 2.0 KB
- agents/debugger.md 2.0 KB
- agents/designer.md 6.4 KB
- agents/executor.md 2.8 KB
- agents/explore.md 2.0 KB
- agents/qa-auditor.md 4.8 KB
- agents/qa-tester.md 3.1 KB
- agents/security-reviewer.md 1.2 KB
- LICENSE 1.0 KB
- reference/db-access.md 2.8 KB
- reference/debugging.md 6.5 KB
- reference/delivery-gate.md 5.8 KB
- reference/domain-context.md 7.8 KB
- reference/domain-rules.md 1.6 KB
- reference/engagement.md 3.8 KB
- reference/functional-ui.md 6.4 KB
- reference/interview.md 6.1 KB
- reference/observability.md 5.0 KB
- reference/plan-grounding.md 4.4 KB
- reference/playwright-cli.md 3.5 KB
- reference/qa-only.md 12 KB
- reference/qa.md 8.2 KB
- reference/research.md 2.7 KB
- reference/role-loop.md 17 KB
- reference/rules.md 982 B
- reference/taste-aesthetics.md 7.5 KB
- reference/taste-skill-v2.md 19 KB
- reference/taste-sources.md 3.0 KB
- reference/ui-ux.md 4.0 KB
- reference/wayfinder.md 7.2 KB
- templates/commit-gate.sh 11 KB runs code
- templates/contrast-gate.mjs 4.3 KB runs code
- templates/db-access/.env.example 451 B
- templates/db-access/check-connection.sh 134 B runs code
- templates/db-access/db-access.cmd 39 B runs code
- templates/db-access/db-access.mjs 5.2 KB runs code
- templates/db-access/read-only-query.sh 138 B runs code
- templates/db-access/schema-summary.sh 132 B runs code
- templates/domain-agent/code-map.md 744 B
- templates/domain-agent/config.json 291 B
- templates/domain-agent/decisions/README.md 270 B
- templates/domain-agent/flows/README.md 800 B
- templates/domain-agent/freshness.md 1.2 KB
- templates/domain-agent/glossary.md 781 B
- templates/domain-agent/index.md 1.3 KB
- templates/domain-agent/invariants.md 504 B
- templates/domain-agent/qa/nav-map.md 1.2 KB
- templates/domain-agent/qa/README.md 830 B
- templates/domain-agent/test-map.md 432 B
- templates/domain-agent/tickets/README.md 204 B
- templates/GOAL.md 1.1 KB
- templates/observability/heartbeat.schema.json 2.5 KB
- templates/observability/sg-emit.sh 5.4 KB runs code
- templates/PLAN.md 1.1 KB
- templates/qa-gate.sh 5.2 KB runs code
- templates/qa-only-gate.sh 5.1 KB runs code
- templates/qa-report.md 2.8 KB
- templates/QA.md 1.1 KB
- templates/R-LOOP.md 569 B
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.
- 2d ago First seen · 71 lines · 24 tokens per session scan A 7bb6c512b5ce
supergoal is a skill published in the GitHub repository cskwork/oh-my-symphony (35 stars, last pushed today), licensed Apache-2.0. It adds 24 tokens to every session and 739 once invoked, about $0.0001 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 2026-10-18.
Other skills, from other repositories
canvas
Read and draw the side-panel canvas with canvasread, canvasedit, or kg canvas. Use when the user mentions a canvas or asks for a diagram.
kagan
Read a ticket, link threads, and move work through the native Kagan board when the user asks for ticket work.
development
An index of programming guidance for Python, Go, Rust, TypeScript, Java, C++, and shell scripting.
bionic-decision-engine
Unified mathematical arbitrator for all resource allocation decisions — money, time, energy, relationships. Absorbs 46 decision protocols + 24 strategy protocols into one dense engine.
video-perception
Use when the user mentions a video file (.mp4, .mov, .avi, .mkv, .webm), a YouTube URL, asks to watch/analyze/review a video, or references video content in conversation.
macos-tools
Native macOS integrations via AppleScript. No API keys needed — works on any Mac.