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.
git clone --depth 1 https://github.com/Obedience-Corp/festivalWrote 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/commands/obedience-corp/festival/festival-plan)<a href="https://agentmods.dev/commands/obedience-corp/festival/festival-plan"><img src="https://agentmods.dev/badge/commands/obedience-corp/festival/festival-plan/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/commands/obedience-corp/festival/festival-plan"><img src="https://agentmods.dev/badge/commands/obedience-corp/festival/festival-plan.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 | $0.00020 | $0.00519 |
| Opus 5 | $0.00010 | $0.00260 |
| Sonnet 5 | $0.00004 | $0.00104 |
| Haiku 4.5 | $0.00002 | $0.00052 |
Grade A, and why
festival-plan 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 10d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan This
The user described work in $ARGUMENTS. Route it and plan it. Do not skip
straight to creating things.
1. Size it out loud
Say which gear this is and why, in one line, before doing anything else.
| Signal | Route |
|---|---|
| One file, one session, reversible | Answer in chat. Create nothing. |
| Linear, a handful of steps, one repo | fest create workflow <name> |
| Multi-phase, multi-repo, needs review gates, or outlives a session | fest create festival |
Most requests are the first row. Reaching for a festival on small work is a failure mode, not thoroughness. If it is genuinely between two rows, pick the smaller one and say it can be promoted later.
2. Show the shape, then ask
Present the proposed phases and the sequences under each. Get a yes before anything is written to disk. You already have what you need to do this, so it costs nothing, and it is the user's one chance to redirect.
3. Plan through the loop
fest create festival --type standard --name <name>
fest next # answer what it asks, repeat
fest validate
Do not scaffold every phase and sequence up front and fill the [REPLACE]
markers afterward. That writes hundreds of unfilled markers, drops
fest validate from 100 to 0, and makes filling them with plausible filler the
score-restoring move. The loop is the planning process.
4. Plan the full scope
Plan the whole thing. Do not trim scope to make the plan look achievable or defer the hard parts to an invented later phase. Scaling down is the user's call, made against a complete plan.
5. Write tutorial-grade tasks
Each task is read by an agent with none of this conversation's context. Name the
files it touches, state error paths as well as the happy path, and cite
file:line anchors you have actually opened. fest validate scores structure,
not substance, so nothing else will catch a thin task.
6. Hand off
End by showing fest status and telling the user how to start execution
(fest next) or hand it to an agent.
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.
- 10d ago First seen · 61 lines · 20 tokens per session scan A 9df6d7d3fa9b
festival-plan is a command published in the GitHub repository Obedience-Corp/festival (49 stars, last pushed today), licensed Apache-2.0. It adds 20 tokens to every session and 519 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.
Other commands, from other repositories
statusbar-style
Switch the status-bar style (classic / capsule / hairline).
sonarqube
Analyze SonarCloud quality issues for a specific PR.
plan-start
5-phase planning command: PRD analysis, design review, technical decisions, dynamic research team, metrics. Produces a complete implementation plan + ADRs before any code is written.
refactor
Analyze code for SOLID violations and suggest targeted improvements.
generate-tests
Generate comprehensive tests for specified code.
superpowers-execute
Execute the current GSD phase plan with Superpowers instead of gsd-execute-phase.