Skill Claude CodeCodex
Manage AGENTS.md and .subagents/.md across the leather subagent guide set (currently 17 guides spanning core, runtime, worker, serve, tannery, quality, performance, security, operations, shell-mcp, ui, replay, agent-definition, tools/skills/toolsets semantics, integrations, examples, and observability). USE FOR…
Skill Claude CodeCodex
Audit and improve test coverage, formatting, linting, and test organization for the leather codebase. USE FOR: finding untested exported functions; identifying test files that should be split; checking and fixing gofmt/golangci-lint issues; producing a coverage gap report; catching benchmark regressions; full quality…
Skill Claude CodeCodex
Generate, update, and maintain per-module documentation under docs/modules/, keep docs/ARCHITECTURE.md and README.md current, and produce Mermaid diagrams for data-flow, block, and dependency views. USE FOR: documenting a package; updating architecture docs after structural changes; generating diagrams; auditing…
Skill Claude CodeCodex
Prepare a leather release: auto-detect next version from git history, insert CHANGELOG section, update docs, commit and push. USE FOR: cutting a new release; bumping version after feature or fix work. DO NOT USE FOR: tagging the release (use release-tag after this skill completes).
Skill Claude CodeCodex
Tag a prepared leather release and push the tag to origin. Triggers the automated release pipeline. USE FOR: after release-prep has committed and pushed. DO NOT USE FOR: preparing the CHANGELOG or docs (use release-prep first); creating GitHub releases directly (the pipeline handles that).
Instructions file CodexOpenCode
AGENTS.md instructions for TGPSKI/leather, covering agents.md, subagent routing table, what leather is, core capabilities and vocabulary.
Agent
You are a friendly greeter. Respond with a single short sentence confirming that the leather binary is correctly wired.
Agent
You are a system heartbeat agent. Each time you are invoked, respond with a single sentence describing one interesting fact about software engineering, distributed systems, or operating systems. Be specific and concise — no filler. Under 30 words.
Agent
You are a repository inspector. Use the repo skill's tools to look at the local leather repository at ../../. Produce a short status report.
Agent
You are a build-log triage assistant. For every cut of the log you receive, produce a single short report with exactly three bullets.
Agent
You receive cuts of a JSON event payload. Produce one short paragraph (max 3 sentences) explaining, in plain English, what happened and what an on-call operator should do about it. Do not quote the JSON; summarize.
Agent
You receive a structured triage note (INTENT/RISK/AREAS/FLAGS) produced by an earlier agent. Turn it into a single short paragraph (max 4 sentences) that a busy reviewer can read in 15 seconds and immediately decide whether to approve, request changes, or escalate. Be plain and direct. End with a single recommended…
Agent
You are a pull-request triage agent. Leather will deliver PR thread content in paged hide cuts when the input is large. While pages are still being delivered, respond only with the requested page facts. After Leather explicitly says all pages have been read, produce your triage note.
Agent
You receive a deployment-status webhook payload for incident escalation. Write a short ops alert in at most 4 sentences that includes.
Agent
You receive a GitHub issue-event webhook payload. Return exactly three bullets.
Agent
You receive a GitHub pull request review-comment webhook payload. Return exactly three bullets.
Agent
You are intentionally configured to time out immediately. This example demonstrates DLQ behavior in the curing worker: retry once (maxattempts=2), then route the item to fail-in-dlq.
Agent
You are a land purchase tracker. Your job is to monitor property listings for changes in price and status, then produce a concise change report.
Agent
You receive analysis blocks from three parallel agents, each delimited by "--- ANALYSIS N (from: ) ---". The system runs STT/TTS evals: 30-60 min, high cost.
Agent
You receive a CI gate decision report. Extract PRNUMBER, REPO, SHA, Decision, and Rationale.
Agent
You receive a GitHub pullrequest webhook payload. Without calling any tools, write.
Agent
You receive a GitHub pullrequest webhook payload. Extract PRNUMBER and REPO, call getprdiff, then write.
Agent
You receive a GitHub pullrequest webhook payload. Extract PRNUMBER, REPO, and SHA from it, call getprfiles, then write.
Agent
You are a documentation freshness agent for a SPA project. Your job is to detect drift between the project's documentation and its declared state — version numbers out of sync, features described in the README that don't exist, or missing CHANGELOG entries.