taskflow

A system for managing work that spans multiple steps, pauses, follow-ups, or delegated tasks.

In plain words
What is it for?
Use it to coordinate long-running jobs, schedule wakeups, track external events, run background cycles, and link delegated work back to a main task.
Why use it?
It keeps goals, tasks, waiting points, and progress connected when work cannot be finished in one conversation.

Skill for Claude CodeCodex

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 skills/codeinfinity1/stram/taskflow
Any agent
npx skills add CodeInfinity1/Stram --skill taskflow
Clone the repo
git clone --depth 1 https://github.com/CodeInfinity1/Stram

Made for: Claude Code, Codex.

Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 600 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 $0.00028 $0.00600
Opus 5 $0.00014 $0.00300
Sonnet 5 $0.00006 $0.00120
Haiku 4.5 $0.00003 $0.00060

Measured yesterday against content hash ccaccc25b4a0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

taskflow 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.

skills/agent-core/taskflow/SKILL.md · 82 lines

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.

TaskFlow

Tool Map

  • cognitive_goal_create
  • cognitive_commitment_record
  • cognitive_commitment_update
  • cognitive_wakeup_schedule
  • cognitive_trigger_record
  • autonomous_task_graph_create
  • autonomous_queue_status
  • autonomous_cycle_run
  • multi_agent_coordinate
  • multi_agent_board

Use this skill when work should outlive one prompt, wait on external events, delegate subtasks, or require resumable progress.

When To Use

  • Multi-step tasks with one owner.
  • Work that needs scheduled follow-up.
  • Work waiting on a person, channel reply, file change, or external system.
  • Background tasks that need status, cancellation, or resumption.
  • Delegated coding or research jobs with child tasks.

Stram Shape

Use cognitive tools and stores for:

  • goals;
  • tasks;
  • wakeups;
  • triggers;
  • runtime event queue;
  • autonomous cycles;
  • memories and learning records.

Workflow

  1. Create or reuse an active goal.
  2. Add tasks with success criteria.
  3. For work that cannot continue now, schedule a wakeup or trigger.
  4. For background progress, run an autonomous cycle.
  5. For delegation, link the child run or worker result back to the goal/task.
  6. On completion, record evidence and learning.

Wait States

Use a waiting state when:

  • credentials are missing;
  • approval is required;
  • a user reply is needed;
  • a remote channel must deliver a response;
  • a scheduled time has not arrived.

Record the reason and the exact next stimulus that should resume the work.

Verification

  • Every durable task should have an owner, status, reason, and evidence refs.
  • Do not create background loops without a stopping condition.
  • Do not mark a task complete merely because a run started.

Safety And Approval

  • Durable goals, commitments, wakeups, triggers, and delegated tasks should be created only when they represent real user intent or model-supported task state.
  • Do not schedule external-visible actions, live sends, installs, deploys, or destructive work without explicit approval.
  • Keep waiting states honest: name the exact blocker, owner, and resume stimulus.
  • Do not mark a task complete until evidence proves the requested end state, not merely a plan, queued run, or worker handoff.

Read the full file on GitHub · 82 lines

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 · 82 lines · 28 tokens per session scan A ccaccc25b4a0

Subscribe to this mod's changes

taskflow is a skill published in the GitHub repository CodeInfinity1/Stram (10 stars, last pushed 22d ago), licensed MIT. It adds 28 tokens to every session and 600 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 2026-08-31.