CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dispatch

Delegate tasks to OpenAI Codex CLI and Google Antigravity CLI from Claude Code. Say 'check with codex' or 'ask gemini for a second opinion' and Claude runs the other agent, keeps a topic-aware session, and critiques the result.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/all-skills/skills/dispatch/SKILL.md
SKILL.md
Quality
Evals
Security

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.

Well-organized and concise, with good illustrative examples and a sensible structure. The main weakness is actionability: the skill describes routing and session behavior conceptually without ever giving the actual commands Claude would run to invoke, resume, or extract results from the external CLIs.

Suggestions

Add an executable command reference: the exact `codex` and `agy` invocations for a new delegation (e.g., prompt passing, sandbox flags, model selection).

Show how sessions are resumed by ID, including how to obtain the ID and what the 'delta bridge' actually consists of.

Include a brief failure-handling note (auth expired, CLI missing, command timeout) so the workflow has at least one explicit checkpoint.

DimensionReasoningScore

Conciseness

The body is short and assumes Claude's competence, but the opening paragraph restates the frontmatter description nearly verbatim — a minor instance of trimming opportunity. No padding explaining known concepts.

4 / 5

Actionability

Some concrete guidance exists (install command, login requirements, model defaults), but the actual CLI invocation commands are absent — '[runs codex review --uncommitted]' and '[resumes the same codex session with a delta bridge]' are illustrative annotations rather than executable instructions, and session/resume mechanics are unspecified.

3 / 5

Workflow Clarity

'What This Skill Does' lists a clear conceptual sequence (route → defaults → synthesize → track), but it stops short of operational steps: no checkpoints, no error handling (e.g., CLI auth failure), and the resume flow is never concretely defined.

3 / 5

Progressive Disclosure

A single compact, self-contained file (~60 lines) with well-organized sections (When to Use, What This Skill Does, Examples, Requirements) and only a one-level external link; nothing here needs a separate reference file.

5 / 5

Total

15

/

20

Passed

Description

87%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.

A strong description: concrete actions, explicit trigger phrases that answer 'when', and a distinct niche with low conflict risk. The only weakness is slightly incomplete keyword coverage for the Antigravity side of the skill.

Suggestions

Add one more natural trigger variation (e.g., 'ask antigravity' or 'get a second opinion from another model') to round out keyword coverage.

Mention session resumption explicitly ('resume a prior thread') so the topic-aware-session capability is fully represented.

DimensionReasoningScore

Specificity

Names the domain ('OpenAI Codex CLI and Google Antigravity CLI') and several concrete actions ('runs the other agent, keeps a topic-aware session, and critiques the result'), but coverage has minor gaps — resuming threads and synthesizing agreement/disagreement are only implied.

4 / 5

Completeness

Explicitly answers both 'what' (delegate/run/session/critique) and 'when' with concrete trigger phrases ('Say "check with codex" or "ask gemini for a second opinion"'), matching the anchor for clear and explicit what-and-when with concrete triggers.

5 / 5

Trigger Term Quality

Includes natural spoken phrases users would say ('check with codex', 'ask gemini for a second opinion') plus tool names ('codex', 'gemini'), but misses a few natural variations such as 'antigravity', 'agy', or 'cross-check'.

4 / 5

Distinctiveness Conflict Risk

A clear niche — running external AI CLIs from inside Claude Code — with distinct tool-name triggers, making conflict with other skills unlikely.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

davepoon/buildwithclaude

Path

plugins/all-skills/skills/dispatch/SKILL.md

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.