
Handoff
Resume the latest agent memory session for the current repo when you open a new chat with no context.
Install
npx skills add https://github.com/rohitg00/agentmemory --skill handoffWhat is this skill?
- User-invocable when you say resume, handoff, or where were we
- Selects sessions via `memory_sessions` MCP with directory-boundary cwd matching (not naive prefix)
- Prefers `completed` sessions over `abandoned` and surfaces unanswered user questions first
- Optional cwd override via `$ARGUMENTS` with normalized absolute paths
- Summarizes title, files touched, decisions, and errors from session observations
Adoption & trust: 1.2k installs on skills.sh; 21.8k GitHub stars; 2/3 security scanners passed (skills.sh audits).
Recommended Skills
Journey fit
Agent session continuity is anchored on agent-tooling because that is where multi-turn coding agents run; the same resume ritual applies in every later phase once you rely on agents. handoff is explicitly about MCP `memory_sessions`, cwd-bound session matching, and narrative recovery—core agent tooling rather than a single app feature.
Common Questions / FAQ
Is Handoff safe to install?
skills.sh reports 2 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.
SKILL.md
READMESKILL.md - Handoff
The user wants to resume work. Optional cwd override: $ARGUMENTS Determine the current project path: if `$ARGUMENTS` is provided, resolve it to an absolute, normalized path (accept relative inputs, e.g. `path.resolve(process.cwd(), $ARGUMENTS)`); otherwise use the current working directory. Call the `memory_sessions` MCP tool. From the result, pick the most recent session whose normalized `cwd` matches the project path with a directory-boundary check — equality OR `session.cwd.startsWith(projectPath + path.sep)` OR `projectPath.startsWith(session.cwd + path.sep)`. Do NOT use a raw string prefix match: it produces false positives across unrelated repos that share a path prefix (e.g. `/repo-a` vs `/repo-a-staging`). Prefer sessions with status `completed` over `abandoned`. If nothing matches, fall back to the single most recent session overall. Once a session is selected: 1. If the session ended on an unanswered user-facing question, surface that question FIRST as the lead. Look for it in `summary` or in the last few observations (type `conversation` with `narrative` ending in `?`). 2. Then summarize the session: title/summary, key files touched, key decisions or errors. Use `memory_recall` with a query derived from the session's top concepts to fetch supporting observations, limit 10. 3. End with a short "next step?" pointer the user can act on. If neither MCP tool is available, fall back to HTTP: `GET $AGENTMEMORY_URL/agentmemory/sessions` and `POST $AGENTMEMORY_URL/agentmemory/recall` with `Authorization: Bearer $AGENTMEMORY_SECRET` when set. Do not invent observations. If the most recent session has zero observations, say so and offer to start fresh.