From halyard
Log completed work to the team knowledge base. Use when finishing a session, completing a task, wrapping up, making a decision, or when prompted to record, document, save, or summarize what was done. Triggers on 'log this', 'write this up', 'save to knowledge base', 'document what we did', 'record this decision', 'summarize this session'. Captures work summaries, architectural decisions, process documentation, and learnings for team visibility.
How this skill is triggered — by the user, by Claude, or both
Slash command
/halyard:log-workThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
After completing all work in this session, call `mcp__plugin_halyard_ask-expert__summarize_work` to log what was accomplished.
After completing all work in this session, call mcp__plugin_halyard_ask-expert__summarize_work to log what was accomplished.
Skip logging if the session was trivial (brief Q&A, simple file reads, casual conversation).
mcp__plugin_halyard_ask-expert__summarize_work(
title: "Implemented user authentication",
summary: "Added JWT-based auth with refresh tokens to the API. Used existing middleware pattern from request-logger. Key decision: chose JWT over sessions because mobile clients need stateless auth.",
entry_type: "WORK_OUTPUT",
tags: ["auth", "api", "security"]
)
| Parameter | Description |
|---|---|
title | Short title for the work entry (required) |
summary | Detailed summary of what was done and why (required) |
entry_type | Type: "WORK_OUTPUT" (default), "DECISION", "PROCESS", or "CONTEXT" |
tags | Tags for categorization |
knowledge_entry_id | ID of an existing entry to update instead of creating new |
source_provider | Source system: "github", "slack", "linear", "claude", "codex", "notion" |
source_url | Navigable link to the source material (PR, ticket, thread) |
source_knowledge_entry_id | ID of another entry this derives from (citation chain) |
supersedes_entry_id | ID of an older entry this replaces (marks it outdated) |
session_id | Link this entry to a specific agent session |
WORK_OUTPUT — Code changes, feature implementations, bug fixes, refactorsDECISION — Architectural choices, design trade-offs, technology selections (include reasoning)PROCESS — How-to documentation, workflow descriptions, setup proceduresCONTEXT — Background research, reference material, exploratory findingsWhen logging work, include 2-4 reflection bullets in your summary covering any of:
Be genuine, not performative. Skip a category if nothing meaningful to say. Be specific — "Grep was faster than Agent for finding the config" beats "tools worked well." If the session was straightforward, a single bullet is fine.
After logging, share one of your reflections with the user as a conversation starter and ask for theirs:
One thing I noticed: [your reflection]. Any thoughts from your side on this session?
source_url when the work relates to a specific PR, ticket, or threadsupersedes_entry_id when your work replaces or updates a previous knowledge entrynpx claudepluginhub halyard-labs/claude-plugin-marketplace --plugin halyardCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.