sammcj/Differed Task Execution
Skills/deferred-task-execution/SKILL.md
Delays execution of a task until a specified time or after a duration. Use when the user asks to run something later, in X minutes/hours, at a specific time, schedule a command, or defer work to a future point.
$ install --global
skillsauthnpx skillsauth add sammcj/agentic-coding Differed Task ExecutionInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 7, 2026, 5:30 AM135.7s2 files scanned
SKILL.md
- name:
- Differed Task Execution
- description:
- Delays execution of a task until a specified time or after a duration. Use when the user asks to run something later, in X minutes/hours, at a specific time, schedule a command, or defer work to a future point.
- allowed-tools:
- [Bash, Read, TaskOutput]
Differed Task Execution
Delays agent work until a user-specified time using a background timer script, then proceeds with the deferred task.
Script Location
~/.claude/skills/deferred-task-execution/scripts/wait-until.sh
Supported Input Formats
The script accepts one argument in either format:
Duration - relative delay from now:
30s,5m,1h,90m,2h30m,1h15m30s
Clock time - specific time of day:
14:30,9:00,3pm,3:30pm- If the time has already passed today, it waits until that time tomorrow
Workflow
Step 1: Confirm the user's intent
Confirm what task to perform and when. Parse their natural language into a wait-until argument:
| User says | Argument |
|-----------|----------|
| "in 30 minutes" | 30m |
| "in an hour" | 1h |
| "in 2 and a half hours" | 2h30m |
| "at 3pm" | 3pm |
| "at 14:30" | 14:30 |
| "at half past 9" | 9:30am |
The script self-handles timezone via the USER_TZ variable defined at the top of wait-until.sh (defaults to Australia/Melbourne). Clock-time arguments are interpreted in that zone regardless of host TZ. Don't ask the user about timezones unless they mention travelling.
Tell the user exactly what will happen and when the timer will fire. Ask them to confirm before starting.
Step 2: Launch the timer in background
Run the wait script using Bash with run_in_background: true:
~/.claude/skills/deferred-task-execution/scripts/wait-until.sh <argument>
Note the task ID from the response -- you need it for the next step.
Step 3: Wait for the timer
Use the TaskOutput tool to block until the timer completes:
- Set
block: true - Set
timeout: 600000(maximum: 10 minutes / 600000ms) - Use the task ID from Step 2
If the wait exceeds 10 minutes: TaskOutput will return before the timer completes because the maximum timeout is 600000ms. Check the output -- if it does not contain "Timer complete", call TaskOutput again with the same task ID and block: true. Repeat until the timer finishes.
Step 4: Execute the deferred task
Once TaskOutput returns output containing "Timer complete. Proceed with deferred task.", carry out whatever work the user requested.
Important Notes
- The agent session must remain open for the timer to work. If the user closes the session, the deferred task will not execute. Warn the user about this.
- For very long waits (multiple hours), remind the user that the session needs to stay active.
- If the user asks to cancel, use
TaskStopwith the background task ID. - Always tell the user the calculated wake time so they know when to expect the action.
Related Skills
sammcj/ml-llm-wiki
development
VerifiedTrustedCommunity
Use when answering questions from this machine-learning knowledge base. Triggers: questions about transformers, attention cost and efficiency, and long-context scaling; 'what do we know about attention', 'check the ML wiki'. Read-only querying of compiled knowledge; to add, update, supersede, lint, or audit, use the llm-wiki skill instead.
136SKILL.mdUpdated Jun 4, 2026
sammcj/llm-wiki
development
VerifiedTrustedCommunity
Use when building or maintaining a self-contained personal knowledge base (an LLM wiki) as plain markdown, optionally opened as an Obsidian vault. Triggers: ingesting sources into a wiki, querying wiki knowledge, linting wiki health, auditing article claims against their sources, superseding stale knowledge, 'add to wiki', or any mention of 'LLM wiki' or 'Karpathy wiki'.
136SKILL.mdUpdated Jun 4, 2026
sammcj/hardware-design
tools
VerifiedTrustedCommunity
Provides guidance and tools for hardware design. Activate when using KiCAD, looking up electronic parts or designing PCBs.
136SKILL.mdUpdated May 29, 2026
sammcj/grill-with-docs
testing
VerifiedTrustedCommunity
Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise.
136SKILL.mdUpdated May 29, 2026