coding-agent
Delegate coding tasks to Codex, Claude Code, or Pi agents via background host sessions. Use when: (1) building or creating new features or apps, (2) reviewing PRs (spawn in temp dir), (3) refactoring large codebases, (4) iterative coding that needs file exploration. NOT for: simple one-liner fixes (just edit), reading code (use read tool), thread-bound ACP harness requests in chat (for example spawn or run Codex or Claude Code in a Discord thread; use sessions_spawn with runtime:"acp"), or any work in ~/clawd workspace (never spawn agents here). Requires OpenClaw host tools with exec_command plus write_stdin.
83
Quality
81%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
85%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that excels at completeness and distinctiveness by clearly defining both positive use cases and explicit exclusions. The main weakness is trigger term quality, which relies heavily on technical jargon (ACP, OpenClaw, exec_command) that users wouldn't naturally use when requesting this functionality.
Suggestions
Add more natural user-facing trigger terms like 'delegate work', 'run in background', 'parallel coding tasks', or 'spawn agent' to improve discoverability
Consider moving technical requirements ('Requires OpenClaw host tools with exec_command plus write_stdin') to a separate field or the skill body, as this is implementation detail rather than selection criteria
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'building or creating new features or apps', 'reviewing PRs', 'refactoring large codebases', 'iterative coding that needs file exploration'. Also specifies what NOT to use it for with concrete examples. | 3 / 3 |
Completeness | Clearly answers both what ('Delegate coding tasks to Codex, Claude Code, or Pi agents via background host sessions') and when with explicit numbered triggers ('Use when: (1) building... (2) reviewing PRs... (3) refactoring... (4) iterative coding'). Also includes explicit NOT for cases which adds clarity. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'Codex', 'Claude Code', 'Pi agents', 'PRs', 'refactoring', but uses technical jargon ('ACP harness', 'OpenClaw host tools', 'exec_command', 'write_stdin') that users wouldn't naturally say. Missing common variations like 'delegate work', 'parallel tasks', 'background coding'. | 2 / 3 |
Distinctiveness Conflict Risk | Very clear niche with distinct triggers around agent delegation and background sessions. The explicit NOT for cases (simple fixes, reading code, ACP harness requests, ~/clawd workspace) actively prevent conflicts with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, actionable skill with excellent executable examples and clear workflow patterns for delegating to coding agents. The PTY requirement is well-emphasized and the parallel execution patterns are particularly useful. Main weaknesses are some redundancy (PTY warnings repeated, learnings section duplicates earlier content) and the monolithic structure that could benefit from splitting into separate files for each agent.
Suggestions
Remove the 'Learnings' section at the end - it repeats information already covered in the main content (PTY, git repo, exec, append_newline)
Consider splitting agent-specific sections (Codex CLI, Claude Code, Pi) into separate reference files with just quick-start examples in the main SKILL.md
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient with good use of tables and code blocks, but includes some redundant explanations (PTY warning repeated multiple times, some verbose commentary like 'it'll read your soul docs'). The learnings section at the end repeats information already covered. | 2 / 3 |
Actionability | Excellent executable examples throughout - copy-paste ready commands with proper flags, real tool parameters documented in tables, and concrete patterns for common workflows like PR reviews and parallel issue fixing. | 3 / 3 |
Workflow Clarity | Clear multi-step sequences with numbered steps for complex operations (parallel issue fixing, PR reviews). Includes monitoring/polling patterns with write_stdin, cleanup steps, and explicit validation points (check if done, kill if needed). | 3 / 3 |
Progressive Disclosure | Content is well-organized with clear sections and a logical flow from quick start to advanced patterns. However, it's a long monolithic file (~200 lines) that could benefit from splitting agent-specific details (Codex, Claude Code, Pi) into separate reference files. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 9 / 11 Passed | |
- Repository
- trpc-group/trpc-agent-go
- Commit
8763418
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.