claude-code

Delegate coding to Claude Code CLI (features, PRs).

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/autonomous-ai-agents/claude-code/SKILL.md

Quality

Content

65%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill is highly actionable with concrete, executable commands and clear orchestration sequences, but it is a verbose monolith with no progressive disclosure and missing validation checkpoints on its batch/dangerous-permission workflows. Splitting reference material into bundle files and adding explicit result-verification steps would substantially improve it.

Suggestions

Move the full CLI flags, slash-command, and keyboard-shortcut tables into reference files (e.g. references/CLI_FLAGS.md, references/SLASH_COMMANDS.md) and link to them from SKILL.md to enable progressive disclosure.

Add explicit validation/verification checkpoints to the Parallel Claude Instances and --dangerously-skip-permissions workflows (e.g. capture each session's final result and confirm success via the JSON `subtype` field before proceeding).

Consolidate the redundant 'Rules for Hermes Agents', 'Pitfalls & Gotchas', and 'Cost & Performance Tips' sections to remove restated guidance and tighten the token budget.

Dimension	Reasoning	Score
Conciseness	The 746-line body is dense and mostly concrete reference material rather than concept-explanation fluff, but it is excessive in volume with redundant sections ('Rules for Hermes Agents' restates 'Pitfalls & Gotchas' and 'Cost & Performance Tips') and could be materially tightened — fitting 'mostly efficient but could be tightened' rather than the noticeably padded anchor below or the lean anchor above.	3 / 5
Actionability	It provides copy-paste-ready executable commands throughout — `terminal(command="claude -p ...")` invocations, tmux send-keys sequences, JSON examples, and jq filters — covering the common print-mode, interactive, PR-review, and parallel-instance cases, matching the fully-executable anchor.	5 / 5
Workflow Clarity	Sequences for print mode, interactive tmux orchestration, and dialog handling are clearly ordered with monitoring checkpoints (capture-pane, look for the ❯ prompt), but the batch 'Parallel Claude Instances' and `--dangerously-skip-permissions` workflows lack explicit output validation/verification steps, so the destructive/batch cap holds this at 3.	3 / 5
Progressive Disclosure	The body is well-structured with clear ## headers and tables, but it is a monolithic 746-line file with zero references to separate files — content that clearly belongs in references (full CLI flag tables, slash-command catalog, keyboard shortcuts, hook/MCP details) is inlined — fitting 'some structure but content that should be separate is inline'.	3 / 5
	Total	14 / 20 Passed

Description

61%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and product-scoped with solid trigger keywords, but its 'when to use' guidance is only weakly implied and it lists just one concrete action. Adding an explicit 'Use when...' clause and one or two more concrete actions would lift completeness and specificity.

Suggestions

Add an explicit trigger clause, e.g. 'Use when delegating coding tasks, opening PRs, or running code review/refactoring via the Claude Code CLI.'

Expand the action list beyond 'Delegate coding' to concrete verbs like 'run one-shot coding tasks, review diffs, and orchestrate multi-turn refactoring sessions'.

Include the full synonym 'pull requests' alongside 'PRs' to broaden natural trigger coverage.

Dimension	Reasoning	Score
Specificity	The phrase 'Delegate coding to Claude Code CLI' names the domain and one concrete action, with '(features, PRs)' adding use-case hints, but coverage is not comprehensive — closer to the anchor naming domain + 1-2 actions than to the multi-action anchor above.	3 / 5
Completeness	It states a clear 'what' ('Delegate coding to Claude Code CLI') but the 'when' is only weakly implied by '(features, PRs)' with no explicit 'Use when...' clause, so per the missing-trigger-guidance cap it stays at 3 rather than 4.	3 / 5
Trigger Term Quality	Natural terms 'coding', 'Claude Code', 'CLI', 'features', and 'PRs' give good keyword coverage with the dominant product-name trigger present, though synonyms like 'pull requests', 'code review', and 'refactoring' are missing — fitting the 'good coverage, a few terms missing' anchor rather than the fully comprehensive one.	4 / 5
Distinctiveness Conflict Risk	Naming the specific product 'Claude Code CLI' gives a clear niche with minimal conflict risk, but 'delegate coding' is a fairly generic action and the related_skills tags (codex, opencode) signal minor overlap with neighboring coding-agent skills, fitting 'mostly distinct; minor overlap risk' rather than the minimal-risk anchor.	4 / 5
	Total	14 / 20 Passed

Validation

75%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (746 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	12 / 16 Passed

Repository: NousResearch/hermes-agent
Path: skills/autonomous-ai-agents/claude-code/SKILL.md
Commit: 6680afb

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.