All skills
simota avatar

/sherpa

@35ffd55
by shingo imotasimota/agent-skills85 stars
15

Guiding workflows by decomposing complex tasks (Epics) into Atomic Steps under 15 minutes each, with progress tracking and drift prevention. Use when complex decomposition is needed.

Use this Skill: https://skilld.dev/gh/simota/agent-skills/sherpa

This session only. Nothing lands on disk.

referenceemergency-protocols.md

≈903 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Emergency Protocols & Base Camp

Purpose: Use this file when work is slipping into emergency conditions, when multi-Epic coordination is needed, or when context-switch cost must be made explicit.

Contents

  • Emergency levels
  • Yellow, Red, and Evacuation protocols
  • Recovery checkpoint
  • Base Camp dashboard
  • Context-switch protocol

Emergency Levels

Level Condition Response
Yellow Alert 1-2 major blockers, falling behind reassess plan, cut scope, or add support
Red Alert critical path blocked, deadline at risk stop and invoke Triage
Evacuation cascading failures, project integrity at risk full stop, stabilize first

Yellow Alert

## Yellow Alert

**Triggered by**:
- velocity ~40% below estimate
- 2 blockers identified
- projected completion ~2x original estimate

### Options
1. Scope Cut (Recommended)
2. Request Reinforcements
3. Extend Timeline
4. Push Through

### Immediate Actions
1. Commit current progress
2. Document blockers
3. Choose the response strategy

Red Alert

## Red Alert

**Triggered by**: critical blocker, cannot proceed

### Emergency Actions
1. Stop dependent work
2. Document current state
3. Commit all progress clearly
4. Communicate to stakeholders
5. Pivot to valid unblocked work if available

### Invoking Triage
/Triage assess current situation
Context: [Epic] blocked by [blocker]
Impact: [N] steps blocked
Need: prioritized recovery options

Evacuation Protocol

## Evacuation Protocol

### Immediate Actions
1. STOP ALL WORK
2. SECURE CURRENT STATE
   git stash
   git status
   git log -5
3. DOCUMENT EVERYTHING
4. INVOKE TRIAGE
5. COMMUNICATE

### Do NOT
- make hasty fixes
- delete anything
- hide the problem
- continue other work until stable

Recovery Checkpoint

## Recovery Checkpoint

**Emergency**: [Description]
**Resolution**: [What fixed it]
**Duration**: [How long it took]

### Post-Mortem Notes
- Root cause
- Warning signs missed
- Prevention

### Plan Adjustments
- [ ] Re-estimate remaining steps
- [ ] Update risk assessments
- [ ] Add monitoring or checks
- [ ] Document in project journal

Base Camp

## Base Camp

**Date**: YYYY-MM-DD
**Session Duration**: Xh Ym
**Total Steps Today**: N completed

### Active Expeditions
| Epic | Progress | Status | Priority | Last Touch |
|------|----------|--------|----------|------------|
| Payment Flow | 40% | Active | P0 | Now |
| User Settings | 25% | Paused | P1 | 2h ago |
| Bug #123 | 0% | Queued | P2 | - |

Priority Matrix

Priority Meaning
P0 - Summit blocking others or due today
P1 - High Camp important this week
P2 - Base Camp useful, flexible timing
P3 - Queue nice to have

Context-Switch Protocol

## Context Switch Request

**Current**: [Epic A] (Step X/Y, Z% complete)
**Target**: [Epic B]

### Switch Assessment
| Factor | Value |
|--------|-------|
| Current step completion | 60% |
| Context switch cost | ~8 minutes |
| Priority comparison | P0 -> P1 |
| Momentum loss | High |

### Recommendation
Finish the current step first when possible.

Context-Switch Cost Reference

These are practical AI-agent context-switch costs. For human deep-focus recovery costs (~23 min), see context-switching-anti-patterns.md.

From -> To Estimated cost Why
Same domain ~3 min context is still warm
Different domain ~8 min mental model rebuild needed
After a long break ~12 min review required

Source: SKILL.md on GitHub

No alerts13d5 checks · Risk SAFE
  • Gen Agent Trust Hub13d

    The skill provides a robust framework for decomposing complex tasks into manageable atomic steps, featuring extensive protocols for risk management and focus protection. No security issues were identified.

  • Socket13d

    No alerts

  • Snyk13d

    Risk: LOW · No issues

  • Runlayer6mo

    1/11 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 35ffd55. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated 2 weeks ago

README badge

README badge for simota/agent-skills/sherpa