メニュー
Fetches GitHub issue context for quest spawning. Parses issue references, retrieves structured data via gh, and produces branch suggestions and PR keywords. Used standalone or as input to quest orchestration.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | missive |
| description | Fetches GitHub issue context for quest spawning. Parses issue references, retrieves structured data via gh, and produces branch suggestions and PR keywords. Used standalone or as input to quest orchestration. |
Fetches structured context from one or more GitHub issues to inform quest planning. Extracts titles, bodies, labels, comments, and produces branch name suggestions and PR close keywords. Designed for both standalone use (display to user) and as input to Gandalf's quest spawning.
/missive #42 or /missive 42 51 67Accepts one or more GitHub issue references as arguments:
#42 or 42#42 #51 #67 or 42 51 67#42 51 #67If no arguments are provided, respond with:
"Usage:
/missive <issue>...— provide one or more issue numbers (e.g.,/missive #42or/missive 42 51 67)."
Extract all issue numbers from the args using the pattern #?\d+. Strip any # prefix to get the raw number.
Read ~/.claude/fellowship.json if it exists. Extract:
branch.pattern — branch name pattern with placeholders (default: null, effective: fellowship/{slug})branch.ticketPattern — regex for ticket extraction (default: [A-Z]+-\d+)issues.autoClose — whether PRs should auto-close issues (default: true)If the config file doesn't exist, use all defaults.
Before fetching, confirm gh is available:
gh --version
If gh is not installed or not authenticated, respond with:
"The
ghCLI is required but not available. Install it from https://cli.github.com/ and authenticate withgh auth login."
For each issue number, run:
gh issue view <number> --json title,body,labels,comments,assignees,milestone
If an issue is not found (non-zero exit), note the error and continue with remaining issues. Do not abort the entire operation for a single missing issue.
For each successfully fetched issue, produce a structured block:
... [truncated] if truncated.Resolve the branch name:
branch.pattern is configured: substitute placeholders per the quest skill's rules, but override {slug} with a slug derived from the issue title (not the task description). The {ticket} placeholder matches branch.ticketPattern against the issue title — if no match, replace with empty string. If {author} is in the pattern but branch.author is not set, replace with empty string. After all substitutions, collapse any resulting double-separators (e.g., // → /, -- → -).fellowship/<number>-<slugified-title> (e.g., fellowship/42-fix-auth-bug). This differs from the quest skill's default (fellowship/{slug}) by including the issue number — linking the branch back to its GitHub issue for traceability.Slug generation: lowercase the issue title, replace spaces with hyphens, strip non-alphanumeric characters (except hyphens), collapse consecutive hyphens, max 50 characters.
issues.autoClose is true (default): Closes #<number>issues.autoClose is false: omit the close keyword, note: "Auto-close disabled — link issue manually."Output one block per issue in this format:
## Issue #<number>: <title>
### Context
**Labels:** <labels>
**Assignees:** <assignees>
**Milestone:** <milestone>
<body text>
**Recent Comments:**
- [@<author> <timestamp>]: <comment text>
- [@<author> <timestamp>]: <comment text>
### Branch
`<resolved branch name>`
### PR Keywords
`Closes #<number>`
If multiple issues were fetched, output all blocks sequentially.
If any issues failed to fetch, append a summary:
### Errors
- Issue #<number>: <error message>
Multi-task orchestrator. Coordinates agent teammates (led by Gandalf) running /quest (code) or /scout (research) workflows. Use when you have multiple independent tasks to run in parallel.
Use for any non-trivial task. Orchestrates the Research-Plan-Implement cycle with compaction between phases, integrating council, lembas, gather-lore, and warden. Enforces discipline and phase gates.
Post-fellowship retrospective analysis. Collects gate history, palantir alerts, and quest metrics to surface patterns and interactively recommend configuration improvements.
Use AFTER writing code and BEFORE submitting a PR. Simulates a strict code reviewer by comparing changes against reference files and documented conventions. Catches "we don't do it that way" feedback before your human reviewer does.
Research & analysis workflow. Investigates questions, gathers findings, optionally validates with a fresh subagent. Produces a report — no code, no PRs, no commits. Use standalone or as a fellowship teammate.
Use at the start of any non-trivial task. Loads focused, task-relevant context by reading CLAUDE.md, scanning for related files, and producing a structured Session Context block. Invoked automatically by quest or standalone via /council.