skills.rest
jeremybrasher
jeremybrasherCOMMUNITY
Explore all skills by this author →
GitHub

show-me-your-work

Maintains an append-only TSV decision log with evidence pointers for long-running or unattended work.

Updated Aug 28, 2026
Search Tags:#autonomous agents#audit trail#decision log#accountability#evidence tracking#tsv#cross-model review
One-click install
npx skills add https://github.com/jeremybrasher/grokbot-skills --skill show-me-your-work-jeremybrasher
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: show-me-your-work
Source: https://github.com/jeremybrasher/grokbot-skills/tree/main/collections/pstack/skills/show-me-your-work
Command: npx skills add https://github.com/jeremybrasher/grokbot-skills --skill show-me-your-work-jeremybrasher

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Long-running, autonomous, or multi-phase work leaves no reviewable trail, so a human who steps away cannot verify what was decided, why, or whether the evidence backs it up. ## Core Features & Use Cases - Structured decision log: One TSV row per decision with timestamp, phase, decision, reason, evidence pointer (commit SHA, PR, file:line), and result. - Append-only audit trail: Wrong calls get superseding rows instead of edits, and the log is audited against the run transcript before handoff. - Cross-model review: A subagent on a different model family reviews the trail and flags weak evidence, skipped verification, or risky choices in an "Attention" section. - Use Case: During an unattended multi-hour refactor, the agent logs each fork, revert, and verification result to decisions.tsv, so a reviewer can follow the evidence pointers and trust the outcome. ## Quick Start Ask the agent to keep a show-me-your-work decision log for this task, recording each decision with its reason, evidence, and result 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 of decisions an AI agent makes?▼

Use a TSV decision log with one row per decision: timestamp, phase, what was decided, why, an evidence pointer like a commit SHA or file:line, and the result. Log forks, completed units, reverts, and blockers rather than every trivial action.

What format works best for an agent decision log?▼

A single TSV file works well because GitHub renders it as a table and `column -s$'\t' -t` renders it in a terminal. Keep cells single-line and make evidence a pointer such as a commit SHA, PR number, or artifact path, never a paragraph.

Should the decision log be committed to git?▼

By default the log is a local working artifact kept at decisions.tsv or .audit/<task-slug>.tsv and left out of git. Commit it only when the work is ambitious enough that a reviewer needs the trail to trust the result.

How do I verify an agent's decision log is accurate?▼

Audit the log against the run's transcript: every row must map to a real action, each evidence pointer must resolve, and unlogged pivots must be added. Then have a subagent on a different model family review the trail and flag weak evidence or risky choices.

When should I not use a decision log?▼

Skip it for short, single-step tasks where the trigger conditions are not met, when required inputs are unavailable, or when a narrower skill fits better. The skill holds rather than guessing when its intake requirements are missing.

Related Skills

Explore More Software Engineering Skills→
Wike-CHIWike-CHI
2

proactive-agent

Persist session-critical decisions and context to SESSION-STATE.md before responding.

Community
Advanced
hanzlahabibhanzlahabib
2

rcode-checkpoint-preview

Guide a human through a five-step code change review before checkpoint approval.

Community
Advanced
cursorcursor
6.6k

show-me-your-work

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

Official
Intermediate
LedgerHQLedgerHQ
615

slack-pr-message

Generates formatted Slack announcement messages for pull requests with package-based prefixes.

Official
Basic