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/reviewgit 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 | $0.00533 |
| Opus 5 | $0.00024 | $0.00267 |
| Sonnet 5 | $0.00010 | $0.00107 |
| Haiku 4.5 | $0.00005 | $0.00053 |
Grade A, and why
review 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.
What it actually says
Context
- Branch: !
git branch --show-current - HEAD: !
git rev-parse --short HEAD
Your job
Run the Review phase of the fable-flow pipeline against the current branch.
Input: $ARGUMENTS (--base REF diff base, default: the base recorded in .fable-flow/plan.md, else the default branch · --reviewers N lens count, default 3 with a plan that has multiple tracks, else 2 · --rounds N max review→fix cycles when --fix, default 2 · --fix apply fixes for actionable findings; without it this command only reports)
- Resolve the base ref. Load
.fable-flow/plan.mdand.fable-flow/task.mdif present — reviewers verify against them; if absent, reviewers verify against the diff's own apparent intent and say so. - Spawn
fable-flow:reviewersubagents in a single message, lenses in priority ordercorrectness,fidelity,integration. Each prompt: the requirements, the plan (if any), the base ref, any relevant lessons from.fable-flow/memory/lessons/, its lens. Save reports to.fable-flow/review-round-<r>-<lens>.md. - Triage: actionable = severity blocker/major with confidence medium+. Everything else is reported but never fix-looped.
Without --fix, the deliverable is the assessment: report the verdicts and findings ranked most-severe first, with each reviewer's execution evidence, and stop — don't change any files.
With --fix: fix actionable findings on the current branch (directly, or via a fable-flow:implementer based on the current SHA for larger fixes), re-run the plan's integration verification, then re-spawn only the lenses that blocked. At most --rounds cycles; whatever survives is reported, not silently retried.
Lead the report with the overall verdict. Quote real command output for any claim that tests pass. If reviewers disagree, present both positions rather than averaging 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.
- 2d ago First seen · 27 lines · 48 tokens per session scan A 7a978e50fc04
review 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 533 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.
hire-assess-fit
Run a culture-add fit assessment for a candidate (NOT culture-fit). Names what the team currently lacks, names what this candidate adds, runs gap-bridge analysis. Refuses generic "great culture fit" framing — culture-fit reproduces, culture-add expands. Used as a SUPPLEMENT to the structured rubric, never as a…
music-suno-prompt
Grounded Suno prompt synthesis from local knowledge corpus + persona canon + label canon. No vibes-prompting.
triage
Triage review findings one by one — present pending TODOs for user decision (approve, skip, modify).
audit-plugin
Audit plugin skills, commands, and agents for structure, size, and naming issues.
earnings-week
Upcoming earnings + dividend events for portfolio holdings over next N weeks (default 2). Surfaces event clusters (multiple positions on same day = concentrated event risk), consensus estimates, key questions to watch. Run before FOMC/earnings-heavy weeks.