From f8d12710f78752bd720e274b600d926574bb647f Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Thu, 2 Jul 2026 11:22:35 +0100 Subject: [PATCH 1/5] Add claude-handoff skill (in-progress) User-invoked skill mirroring handoff, but hands the conversation summary to a fresh background agent via `claude --bg` instead of saving a markdown document. Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/claude-handoff/SKILL.md | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) create mode 100644 skills/in-progress/claude-handoff/SKILL.md diff --git a/skills/in-progress/claude-handoff/SKILL.md b/skills/in-progress/claude-handoff/SKILL.md new file mode 100644 index 0000000..0e25b18 --- /dev/null +++ b/skills/in-progress/claude-handoff/SKILL.md @@ -0,0 +1,18 @@ +--- +name: claude-handoff +description: Hand off the current conversation to a fresh background agent that picks up the work immediately. +argument-hint: "What will the next session be used for?" +disable-model-invocation: true +--- + +Compose a handoff summarising the current conversation so a fresh agent can continue the work, then launch it as a background agent seeded with that summary — do not save a document. + +Launch with `claude --bg ""`, passing the summary as the prompt. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. + +Include a "suggested skills" section in the summary, instructing the agent which skills to invoke. + +Do not duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead. + +Redact any sensitive information, such as API keys, passwords, or personally identifiable information — it becomes the agent's prompt. + +If the user passed arguments, treat them as a description of what the next session will focus on and tailor the summary accordingly. From 8f3401d1a17ab65e92b18a6ad9d301e9addf135f Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Thu, 2 Jul 2026 11:24:51 +0100 Subject: [PATCH 2/5] claude-handoff: reuse original handoff wording Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/claude-handoff/SKILL.md | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/skills/in-progress/claude-handoff/SKILL.md b/skills/in-progress/claude-handoff/SKILL.md index 0e25b18..b326910 100644 --- a/skills/in-progress/claude-handoff/SKILL.md +++ b/skills/in-progress/claude-handoff/SKILL.md @@ -1,18 +1,16 @@ --- name: claude-handoff -description: Hand off the current conversation to a fresh background agent that picks up the work immediately. +description: Hand the current conversation off to a fresh background agent that picks up the work immediately. argument-hint: "What will the next session be used for?" disable-model-invocation: true --- -Compose a handoff summarising the current conversation so a fresh agent can continue the work, then launch it as a background agent seeded with that summary — do not save a document. +Write a handoff document summarising the current conversation so a fresh agent can continue the work. Instead of saving it, launch a background agent seeded with the document as its prompt: `claude --bg ""`. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. -Launch with `claude --bg ""`, passing the summary as the prompt. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. - -Include a "suggested skills" section in the summary, instructing the agent which skills to invoke. +Include a "suggested skills" section in the document, which suggests skills that the agent should invoke. Do not duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead. -Redact any sensitive information, such as API keys, passwords, or personally identifiable information — it becomes the agent's prompt. +Redact any sensitive information, such as API keys, passwords, or personally identifiable information — the document becomes the agent's prompt. -If the user passed arguments, treat them as a description of what the next session will focus on and tailor the summary accordingly. +If the user passed arguments, treat them as a description of what the next session will focus on and tailor the doc accordingly. From 8347a416143521d6ead6ce7d9cd35f2d1339fa0a Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Thu, 2 Jul 2026 11:25:18 +0100 Subject: [PATCH 3/5] claude-handoff: require descriptive --name Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/claude-handoff/SKILL.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/skills/in-progress/claude-handoff/SKILL.md b/skills/in-progress/claude-handoff/SKILL.md index b326910..a68e6b5 100644 --- a/skills/in-progress/claude-handoff/SKILL.md +++ b/skills/in-progress/claude-handoff/SKILL.md @@ -5,7 +5,9 @@ argument-hint: "What will the next session be used for?" disable-model-invocation: true --- -Write a handoff document summarising the current conversation so a fresh agent can continue the work. Instead of saving it, launch a background agent seeded with the document as its prompt: `claude --bg ""`. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. +Write a handoff document summarising the current conversation so a fresh agent can continue the work. Instead of saving it, launch a background agent seeded with the document as its prompt: `claude --bg --name "" ""`. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. + +Always pass `-n`/`--name` with a descriptive name (e.g. `--name "Fix login bug"`) — it sets the display name shown in the job list, session picker, and terminal title. Include a "suggested skills" section in the document, which suggests skills that the agent should invoke. From 5a7ee93f0129794acbb2950f2e9636e49b29f71a Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Thu, 2 Jul 2026 11:26:55 +0100 Subject: [PATCH 4/5] claude-handoff: say handoff summary, not document Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/claude-handoff/SKILL.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/skills/in-progress/claude-handoff/SKILL.md b/skills/in-progress/claude-handoff/SKILL.md index a68e6b5..88ce8be 100644 --- a/skills/in-progress/claude-handoff/SKILL.md +++ b/skills/in-progress/claude-handoff/SKILL.md @@ -5,14 +5,14 @@ argument-hint: "What will the next session be used for?" disable-model-invocation: true --- -Write a handoff document summarising the current conversation so a fresh agent can continue the work. Instead of saving it, launch a background agent seeded with the document as its prompt: `claude --bg --name "" ""`. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. +Write a handoff summary of the current conversation so a fresh agent can continue the work. Instead of saving it, launch a background agent seeded with the summary as its prompt: `claude --bg --name "" ""`. It starts in the current working directory and returns immediately; the user manages it with `claude agents`. Always pass `-n`/`--name` with a descriptive name (e.g. `--name "Fix login bug"`) — it sets the display name shown in the job list, session picker, and terminal title. -Include a "suggested skills" section in the document, which suggests skills that the agent should invoke. +Include a "suggested skills" section in the summary, which suggests skills that the agent should invoke. Do not duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead. -Redact any sensitive information, such as API keys, passwords, or personally identifiable information — the document becomes the agent's prompt. +Redact any sensitive information, such as API keys, passwords, or personally identifiable information — the summary becomes the agent's prompt. -If the user passed arguments, treat them as a description of what the next session will focus on and tailor the doc accordingly. +If the user passed arguments, treat them as a description of what the next session will focus on and tailor the summary accordingly. From b25e2b3787334a30bf0b1afbd9204db311bf426e Mon Sep 17 00:00:00 2001 From: Matt Pocock Date: Thu, 2 Jul 2026 11:32:04 +0100 Subject: [PATCH 5/5] claude-handoff: add to in-progress README index Co-Authored-By: Claude Opus 4.8 (1M context) --- skills/in-progress/README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/skills/in-progress/README.md b/skills/in-progress/README.md index b283f5b..8b91783 100644 --- a/skills/in-progress/README.md +++ b/skills/in-progress/README.md @@ -8,3 +8,4 @@ Skills that are still being developed. They're not ready to ship — expect roug - **[writing-beats](./writing-beats/SKILL.md)** — Shape an article as a journey of beats, choose-your-own-adventure style. Pick a starting beat, write only that beat, then pivot to the next, until the article reaches a natural end. - **[writing-fragments](./writing-fragments/SKILL.md)** — Grilling session that mines you for fragments — heterogeneous nuggets of writing — and appends them to a single document as raw material for a future article. - **[writing-shape](./writing-shape/SKILL.md)** — Take a markdown file of raw material and shape it into an article paragraph by paragraph, arguing format choices at each step. +- **[claude-handoff](./claude-handoff/SKILL.md)** — Hand the current conversation off to a fresh background agent that picks up the work immediately, seeded with a handoff summary via `claude --bg`. User-invoked.