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/thierryn/fire-flow/fire-3-executegit clone --depth 1 https://github.com/ThierryN/fire-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 | $0.00011 | $0.24511 |
| Opus 5 | $0.00005 | $0.12256 |
| Sonnet 5 | $0.00002 | $0.04902 |
| Haiku 4.5 | $0.00001 | $0.02451 |
Grade A, and why
fire-3-execute 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 — 2,047 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/fire-3-execute
Execute phase N with breath-based parallelization and honesty protocols
Purpose
Execute all plans in a specified phase using breath-based parallel execution. Each plan is handled by a fire-executor agent that applies honesty protocols, references skills, and creates unified handoff documents. After execution completes, a fire-verifier validates the work.
Arguments
arguments:
phase_number:
required: true
type: integer
description: "Phase number to execute (e.g., 1, 2, 3)"
example: "/fire-3-execute 3"
optional_flags:
--breath: "Execute only a specific breath (e.g., --breath 2)"
--plan: "Execute only a specific plan (e.g., --plan 03-02)"
--skip-verify: "Skip verification after execution (not recommended)"
--continue: "Continue from last checkpoint (for interrupted execution)"
--auto-continue: "Enable Double-Shot Latte pattern - no 'continue?' interrupts"
--skip-review: "Skip the parallel code review (not recommended)"
--autonomous: "Auto-route merge gate verdicts without human checkpoints (DEFAULT in v10.0, used by /fire-autonomous)"
--manual: "Opt-in to human checkpoints at merge gate (v9.0 behavior)"
--worktree: "Use git worktree isolation for parallel breath execution (v10.0)"
--model-split: "Force architect/editor model split for all tasks (v10.0)"
--token-efficient: "Enable ALAS context slicing for reduced token usage (~60% reduction). Default: OFF (full context for best quality). Toggle ON for budget-conscious runs or large phase counts."
Auto-Continuation Mode (Double-Shot Latte Pattern)
When --auto-continue is enabled, execution proceeds without "continue?" interrupts:
┌─────────────────────────────────────────────────────────────────────────────â”
│ AUTO-CONTINUATION ACTIVE │
├─────────────────────────────────────────────────────────────────────────────┤
│ │
│ Breath completion → Automatic checkpoint → Next breath starts immediately │
│ │
│ Benefits: │
│ • Uninterrupted execution flow │
│ • Faster phase completion │
│ • Progress saved at each breath checkpoint │
│ │
│ Safety: │
│ • Checkpoints created between breaths │
│ • Can resume from any breath if interrupted │
│ • Blocking errors still pause execution │
│ │
└─────────────────────────────────────────────────────────────────────────────┘
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 · 2,047 lines · 11 tokens per session scan A 70c7e7dede1a
fire-3-execute is a command published in the GitHub repository ThierryN/fire-flow (77 stars, last pushed 20d ago), licensed MIT. It adds 11 tokens to every session and 24,511 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
claude-flow-swarm
Coordinate multi-agent swarms for complex tasks.
bootstrap
PACT session-start ritual — identify the session team (platform-provisioned), secretary spawn, paused-state surface, bootstrap marker.
ox-session-review
Command "ox-session-review" from sageox/ox, covering failure-mode watch-list (read first), from the ledger root. should print 0, phase 1 — scan & score (read-only), quality buckets (first match wins) and removal candidates.
add-agent
引导新增一个 Agent 适配器。用法 /add-agent.
handoff
Package current state for the next agent.
xdo
Direct development by the Main Agent.