From Allura Scout
Use before planning, building, reviewing, or handing off work when token cost or context drift matters. Produces compact evidence-backed ContextPackets from local files, Allura memory, and relevant docs. Read-only by default.
How this skill is triggered — by the user, by Claude, or both
Slash command
/allura-scout:allura-scoutThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use Allura Scout to reduce token cost before expensive agent work.
Use Allura Scout to reduce token cost before expensive agent work.
Use this skill when:
Scout is read-only by default.
Do not edit files, mutate databases, or mark work done. The only allowed write is an optional closeout receipt after the packet is produced.
ContextPacket under the token budget.Default target: 5,000 context tokens. Hard ceiling: 8,000 context tokens unless the user explicitly approves more.
Prefer:
Avoid:
goal: "..."
summary: "..."
relevant_files:
- path: "..."
reason: "..."
relevant_memories:
- id: "..."
reason: "..."
decisions:
- "..."
risks:
- "..."
token_budget:
target_context_tokens: 5000
hard_limit_tokens: 8000
recommended_route:
next_agent: "builder | reviewer | architect | product | validation | stop"
why: "..."
Public users see Allura Scout as a context-reduction plugin. Internal Team RAM role names are implementation details, not public product language.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub allura-ecosystem/allura_memory --plugin allura-scout