show-me-your-work

Maintains an append-only TSV decision log recording choices, evidence, and results for reviewable work.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/gmackie/agent-skills --skill show-me-your-work-gmackie
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: show-me-your-work
Source: https://github.com/gmackie/agent-skills/tree/main/skills/show-me-your-work
Command: npx skills add https://github.com/gmackie/agent-skills --skill show-me-your-work-gmackie

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Long-running or unattended agent work leaves reviewers unable to reconstruct what was decided, why, and on what evidence without rereading entire transcripts. This Skill keeps a single canonical decision trail so a human can audit the result after stepping away. ## Core Features & Use Cases - Structured TSV decision log: One row per decision with timestamp, phase, decision, why, evidence, and result columns, rendered as a sortable table on GitHub or in a terminal. - Safe logging helper: The scripts/log.sh utility stamps timestamps, writes headers, strips stray tabs and newlines, and escapes spreadsheet formula characters in untrusted cells. - End-of-run auditing: Walks the log against the run transcript and spawns a cross-model subagent review, ending each reply with an Attention section of flagged risks. - Use Case: During a multi-week migration, log each fork, revert, and verification result to decisions.tsv, then commit the log so the PR reviewer can follow evidence pointers instead of trusting a summary. ## Quick Start Ask the agent to keep a show-me-your-work decision log for this task and record each major decision with its evidence in a TSV file.

Frequently Asked Questions about show-me-your-work

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I keep an audit trail for autonomous agent work?▼

Keep a single TSV decision log with one row per decision recording timestamp, phase, decision, reason, evidence pointer, and result. Append rows at decision points and checkpoints using the log.sh helper, then audit the log against the run transcript before handing back.

What format works best for a decision log reviewers can read?▼

TSV works well because GitHub renders it as a sortable table, spreadsheets open it directly, and terminal tools like column display it cleanly. Each row stays single-line with evidence as a pointer such as a commit SHA, PR number, or file:line reference.

Should the decision log be committed to git?▼

By default the log stays local as a working artifact at decisions.tsv or under .audit/ and is left out of git. Commit it only when the work is ambitious enough that a reviewer needs the trail to trust the result, such as a large port or multi-week migration.

Why escape spreadsheet formula characters in log cells?▼

Cells starting with =, +, -, or @ can trigger formula execution when a reviewer opens the TSV in a spreadsheet. The log.sh helper prefixes such cells with a single quote so attacker-controlled text like PR titles or filenames cannot execute.

When should I not log a row in the decision trail?▼

Skip trivial and self-evident actions; log only decision points and checkpoints such as a fork chosen, a completed unit with its verification result, a pivot or revert, or a blocker. If nobody would audit a row, it does not earn its place.