Select facts, artifacts, Layer 3 Skills, and References when decision context is stale, redundant, uncertain, or over budget; skip transfer and sufficient current context.
Scanned 9/8/2026
Install to Claude Code
npx -y skills add machenjie/rd-skills --skill task-context-selection --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Task Context Selection?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/machenjie-task-context-selection)More formats (shields.io, HTML) on the badges page.
---
name: task-context-selection
description: "Select facts, artifacts, Layer 3 Skills, and References when decision context is stale, redundant, uncertain, or over budget; skip transfer and sufficient current context."
---
# task-context-selection
## Registry Trigger
**Use when**
- a working decision needs a minimum set of current facts, artifacts, Layer 3 Skills, or References
- candidate context is stale, irrelevant, redundant, uncertain, or exceeds the available budget
**Do not use when**
- current non-redundant context already supports the decision with no material omission
- context must be packaged or transferred downstream
## Skill Role
Select working context before or during one decision. Do not define routing,
phases, permissions, or downstream transfer; `task-handoff-context` owns
packaging context for another agent.
## Inputs
- the exact current decision, alternatives, and failure boundary
- candidate facts and artifacts with source identity and freshness basis
- candidate Layer 3 Skills and References, plus the available context budget
## High-Value Rules
- Include only context that can distinguish an alternative, constraint, or failure boundary in the current decision.
- Record source identity, freshness basis, and decision use for every included fact or artifact.
- Select Layer 3 Skills and References only when their material supplies a non-redundant fact, method, constraint, or comparison used by the decision.
- Exclude irrelevant, stale, and redundant context. Record omissions that could change the decision.
- Refresh an included item when a material state change can invalidate its freshness basis or decision use.
- Make the context-budget tradeoff explicit: preserve authoritative evidence and compress or drop replaceable narrative.
- Report uncertainty and residual risk without filling gaps with plausible context.
## Anti-Patterns
- More context does not make an unsupported fact reliable.
- A recent timestamp alone does not establish freshness for the current state.
- Familiar material without a named decision use consumes budget without reducing uncertainty.
- Budget pressure does not justify dropping source identity or a material omission.
## Execution Checklist
1. Name the current decision and the distinctions it must resolve.
2. Compare candidate facts and artifacts by source, freshness, and decision use.
3. Select the minimum non-redundant facts, artifacts, Layer 3 Skills, and References.
4. Record excluded context, omissions, refresh triggers, and the context-budget tradeoff.
5. Return remaining uncertainty and residual risk.
## Stop Conditions
- Return `insufficient` when a material fact lacks source identity or a defensible freshness basis.
- Stop selection when the available budget cannot retain the minimum evidence needed for the decision; name the displaced evidence and tradeoff.
## Output Contract
- working-context selection with current decision, selected facts and artifacts, source, freshness, decision use, selected Layer 3 Skills and References, excluded context, refresh triggers, omissions, uncertainty, context-budget tradeoff, and residual risk
## Targeted References
| Path | Type | Load when | Do not load when | Required by | Required output |
|---|---|---|---|---|---|
| [context selection](references/context-selection-checklist.md) | decision-checklist | candidate facts artifacts Layer 3 Skills or References need source freshness decision-use or budget comparison | the current minimum set and refresh conditions are already explicit | analysis-agent, task-agent, review-agent | checklist-result, residual-risk |
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!