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/shyftlabs/continuum/pre-taskgit clone --depth 1 https://github.com/shyftlabs/continuumWhat 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 | $0.00000 | $0.00530 |
| Opus 5 | $0.00000 | $0.00265 |
| Sonnet 5 | $0.00000 | $0.00106 |
| Haiku 4.5 | $0.00000 | $0.00053 |
Grade A, and why
pre-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 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.
This is a copy
100% identical to pre-task — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hook pre-task
Execute pre-task preparations and context loading.
Usage
npx claude-flow hook pre-task [options]
Options
--description, -d <text>- Task description for context--auto-spawn-agents- Automatically spawn required agents (default: true)--load-memory- Load relevant memory from previous sessions--optimize-topology- Select optimal swarm topology--estimate-complexity- Analyze task complexity
Examples
Basic pre-task hook
npx claude-flow hook pre-task --description "Implement user authentication"
With memory loading
npx claude-flow hook pre-task -d "Continue API development" --load-memory
Manual agent control
npx claude-flow hook pre-task -d "Debug issue #123" --auto-spawn-agents false
Full optimization
npx claude-flow hook pre-task -d "Refactor codebase" --optimize-topology --estimate-complexity
Features
Auto Agent Assignment
- Analyzes task requirements
- Determines needed agent types
- Spawns agents automatically
- Configures agent parameters
Memory Loading
- Retrieves relevant past decisions
- Loads previous task contexts
- Restores agent configurations
- Maintains continuity
Topology Optimization
- Analyzes task structure
- Selects best swarm topology
- Configures communication patterns
- Optimizes for performance
Complexity Estimation
- Evaluates task difficulty
- Estimates time requirements
- Suggests agent count
- Identifies dependencies
Integration
This hook is automatically called by Claude Code when:
- Starting a new task
- Resuming work after a break
- Switching between projects
- Beginning complex operations
Manual usage in agents:
# In agent coordination
npx claude-flow hook pre-task --description "Your task here"
Output
Returns JSON with:
{
"continue": true,
"topology": "hierarchical",
"agentsSpawned": 5,
"complexity": "medium",
"estimatedMinutes": 30,
"memoryLoaded": true
}
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 · 112 lines · 0 tokens per session scan A 93597737b83e
pre-task is a command published in the GitHub repository shyftlabs/continuum (84 stars, last pushed 11d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 530 tokens. A static security scan graded it A with 0 findings. It is 100% identical to pre-task, differing in 0 lines, and is treated as a copy.
Other commands, from other repositories
port-commits
把已落地、可追溯的一组 commit，用 cherry-pick（最稳妥） 同步到目标分支（常见：main → hc-0730）。.
auto-goal
goal 래퍼 — /goal 생성, 상태 확인, 완료/blocked handoff를 goal tool 또는 slash command로 연결합니다.
auto-plan
SPEC 작성 — 코드베이스 분석 후 EARS 요구사항, 구현 계획, 인수 기준을 생성합니다.
auto-verify
프론트엔드 UX 검증 — Playwright 기반 비주얼 검증을 실행합니다.
doctor
Diagnose installation health. Check Node, CDP bridge, rn-fast-runner (iOS), rn-android-runner (Android), maestro-runner, simulators, Metro, CDP, injected helpers, ffmpeg, physical devices, plugin version, Vercel rules sync. Reports what's missing — does NOT modify your project.
market-radar
Proactive external competitive / market intelligence scan. Routes to market-radar agent. Default = periodic sweep; focused = topic / market / competitor deep-dive.