dvy1987/agent-loom / .agents/skills/memory-handoff
memory-handoff
VerifiedWrite concise next-agent handoff summaries across sessions, tools, and coding agents. Load when the user says handoff, next agent should know, save context, summarize where we are, switching agents, before ending a meaningful session, or when the user asks to commit, push, commit and push, create a git commit, push to origin, or publish commits — commit/push requests MUST run this skill first to prepare handoff docs, then proceed with git operations.
Memory Handoff
You preserve continuity for the next agent. A handoff is short, actionable, and focused on what would otherwise be lost.
Trigger Policy
Run when a future agent would lose important context:
- Meaningful code changes, debugging discoveries, architecture debates, spec changes, or deferred decisions.
- End of a long session with unresolved work.
- Before switching agents or tools.
- User says "handoff", "summarize where we are", "save context", "memory handoff", or "next agent should know".
- User asks to commit and/or push ("commit", "create a commit", "commit these changes", "prepare commit", "push", "push to origin", "git push", "commit and push", "commit and push when ready") — run full handoff workflow before staging/committing/pushing so the next session has continuity. Pair with
git-workflow-and-versioningfor git operations after handoff is saved.
Do not run after trivial interactions.
Workflow
- Read
docs/memory/project-index.mdand latestdocs/memory/agent-handoffs.mdif present. - Inspect current session context and
git status --short. - Summarize only durable context: done, debated, decisions, blockers, deferred items, next steps, revisit triggers.
- Append the handoff to
docs/memory/agent-handoffs.md. - Update
docs/memory/current-state.mdif the project state changed. - Update
docs/memory/project-index.mdwith the handoff entry. - Update knowledge graph — run
python3 .agents/skills/knowledge-graph/scripts/build_graph.py --incremental. If it fails, add### Graphnote in handoff; do not block save. - Append changes to
docs/skill-outputs/SKILL-OUTPUTS.md. - Optional harness mining: When session had repeated agent failures and harness evolution is queued, distill failure digests per
references/harness-trajectory-mining.md— never raw transcripts.
Template
## YYYY-MM-DD HH:MM - Handoff
### Done
- <completed work>
### Debated
- <tradeoff and conclusion>
### Decisions
- <decision and link/reference>
### Deferred
- <parked item and why>
### Next Agent Should Know
- <highest-value continuity note>
### Revisit Triggers
- <conditions that reopen decisions>
### Working Tree
- <clean or relevant dirty files>
Hard Rules
- Keep each handoff under 80 lines.
- Do not include secrets, tokens, or raw private data.
- Link to decision entries instead of repeating long rationale.
- If the handoff gets repetitive, call
memory-compact.
Example
User: "I'm moving this to another agent, save a handoff."
Output: append a timestamped handoff with current status, unresolved tasks, and files touched.
Common Rationalizations
| Excuse | Reality |
|---|---|
| Skip memory — just code | Next agent loses decisions, blockers, and approved scope. |
| Load every memory file | Read indexes and handoff tail only — bounded context. |
| Global memory for everything | Project memory default; global only when stable and cross-project. |
| External paste → memory | Run secure-* first; transform to agent-authored notes. |
Verification
- Correct sub-skill routed with reason
- No secrets or raw transcripts persisted
- Files changed listed in Impact Report
- Security gate noted when external content involved
Red Flags
- Handoff exceeds 80-line budget
- Secrets tokens or raw private data in handoff body
- Long decision rationale pasted instead of log link
- Git state omitted from handoff next-agent context
Prune Log
Last pruned: 2026-07-05
- Deep learn-from: harness-trajectory-mining.md for RHO prerequisite path
Impact Report
After completing, report:
Handoff saved: docs/memory/agent-handoffs.md
Current state updated: yes/no
Index updated: yes/no
Next recommended action: <one sentence>
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 466k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 196k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |