flow-task

A command adapter for the canonical flow/task lifecycle, a structured process for planning an exploration task in Claude Code.

In plain words
What is it for?
Starting an exploration task, presenting the allowed choices, following the flow-planning procedure, and capturing completion evidence.
Why use it?
It ensures the task follows the designated planning procedure and is not complete until its worksheet and supporting evidence are recorded.

Command

Install

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.

agentmods
npx agentmods add commands/cofin/flow/flow-task
Clone the repo
git clone --depth 1 https://github.com/cofin/flow
Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 471 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00008 $0.00471
Opus 5 $0.00004 $0.00235
Sonnet 5 $0.00002 $0.00094
Haiku 4.5 $0.00001 $0.00047

Measured yesterday against content hash 789d149bd23b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

flow-task 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.

commands/flow-task.md · 65 lines

What it actually says

{
  "agent": "executor",
  "argument_schema": {
    "optional": [],
    "required": [
      "exploration"
    ],
    "syntax": "<exploration>"
  },
  "bounds_enforcement": "agent_validated",
  "canonical_id": "flow/task",
  "capability_evidence": "Claude Code declared AskUserQuestion contract",
  "choice_max": 4,
  "choice_min": 2,
  "completion_gates": [
    "worksheet_complete",
    "evidence_captured"
  ],
  "custom_answer_behavior": "native_custom_input",
  "disabled_choice_policy": "omit",
  "fallback": "Use Flow to create an exploration task",
  "git_tags": "forbidden",
  "host": "claude_code",
  "instruction": "Load the lifecycle owner and follow the canonical procedure source directly.",
  "interaction_mode": "none",
  "invocation": "/flow-task",
  "kind": "flow_command_adapter",
  "lifecycle_owner": "flow-planning",
  "multi_select": true,
  "mutability": "planning_write",
  "mutual_exclusion": true,
  "plan_capability": "preferred",
  "procedure_source": "skills/flow/references/task.md",
  "question_capability": null,
  "question_permission_check": "declared_and_allowed",
  "question_tool": "AskUserQuestion",
  "question_transport": "conditional_native",
  "runtime_dependency": "agent_file_tools_only",
  "sequential_fallback": true,
  "shared_contracts": [
    "flow-state-v1",
    "worksheet-execution-v1"
  ],
  "state_operations": [
    "create",
    "claim",
    "discover",
    "release",
    "close"
  ],
  "supported_selection_modes": [
    "binary",
    "single_select",
    "multi_select"
  ]
}
Changes

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.

  1. yesterday First seen · 65 lines · 8 tokens per session scan A 789d149bd23b

Subscribe to this mod's changes

flow-task is a command published in the GitHub repository cofin/flow (16 stars, last pushed 2d ago), licensed Apache-2.0. It adds 8 tokens to every session and 471 once invoked, about $0.0000 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.