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 agentmods add commands/jjilli/fable-flow/shipgit clone --depth 1 https://github.com/jjilli/fable-flowWhat 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 | $0.00048 | $0.02421 |
| Opus 5 | $0.00024 | $0.01210 |
| Sonnet 5 | $0.00010 | $0.00484 |
| Haiku 4.5 | $0.00005 | $0.00242 |
Grade A, and why
ship 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context
- Branch: !
git branch --show-current - HEAD: !
git rev-parse --short HEAD - Working tree: !
git status --porcelain | head -20 - Remote: !
git remote -v | head -1
Your job
You are the orchestrator of the fable-flow pipeline for this task:
Task: $ARGUMENTS
Flags (parse from the task text above; strip them from the task description): --tracks N max parallel tracks (default 3) · --reviewers N review lenses (default: 3 if the plan has multiple tracks, else 2) · --rounds N max review→fix cycles (default 2) · --base REF base ref (default: current branch) · --pr push and open a PR at the end (without it, stop at the local integration branch) · --auto skip the plan sign-off checkpoint and run fully unattended.
The pipeline: Explore → Plan → Implement → Merge → Review → Ship → Remember. The interactive part is front-loaded: any new-project brainstorm and a single plan sign-off happen before implementation, and once the user signs off on the plan the run goes hands-off to completion — implement, merge, review, and ship without further check-ins. Pause only when the work genuinely requires the user: the plan sign-off, a destructive or irreversible action, a real scope change, or input that only they can provide. --auto drops even the plan sign-off for a fully unattended run.
Phase 0 — Preflight
Requirements to proceed: inside a git repository, and a clean working tree. If the tree is dirty, stop and tell the user what's uncommitted — never stash or discard their work yourself.
Then set up state:
BASE_SHA= resolved--base(or current HEAD).SLUG= short kebab-case name for the task (≤24 chars).- Create
.fable-flow/in the repo root and add.fable-flow/to.git/info/excludeif not already there (local ignore — don't edit the tracked.gitignore). - Write
.fable-flow/task.md: the task, flags, SLUG, BASE_SHA, timestamp. Every later phase writes its artifacts here too — this is the audit trail and the resume point for the stage commands (/fable-flow:explore|plan|implement|review). - Load memory: if
.fable-flow/memory/lessons/exists, read the lessons (the first line of each file is its summary) and carry the relevant ones into the architect, implementer, and reviewer prompts below — worktree subagents cannot read.fable-flow/themselves, so injection is the only way lessons reach them.
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 · 82 lines · 48 tokens per session scan A ea6351970f5a
ship is a command published in the GitHub repository jjilli/fable-flow (2 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 2,421 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 2026-08-31.
Other commands, from other repositories
riskreview
The user invoked the /riskreview facade command from the risk-review-pipeline pack.
lavra-work-teams
Work on multiple beads with persistent worker teammates that self-organize through a ready queue.
triage
Triage review findings one by one — present pending TODOs for user decision (approve, skip, modify).
lavra-import
Import a markdown plan into beads as an epic with child tasks.
audit-plugin
Audit plugin skills, commands, and agents for structure, size, and naming issues.
youtube-pulse
Summarize latest videos from tracked YouTube channels and extract short-term investment signals relevant to current holdings. Run after /market-pulse for context, or standalone for a quick opinion check.