invoking-codex-exec

Use when delegating a single coding task to `codex exec` ("hand off to codex", "run codex on this", "dispatch codex on this ticket", any one-shot invocation). Covers flags, sandbox traps, monitoring, and recovery. Not for multi-issue parallel batches — use codex-issue-waves for those.

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%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.

A high-quality, operationally precise skill body: executable commands, clear role-based workflows with validation and recovery loops, and a real referenced helper script. The main weaknesses are minor restated material and a sizable inline contract/enforcement section that could be split into a reference file.

Suggestions

Consolidate the repeated "claude is the manager, codex is the worker" model and the "don't use --full-auto" warning into a single authoritative section to remove restatements.

Move the full reviewer JSON output contract and the orchestrator-side read-only enforcement script into a separate reference file (e.g. references/reviewer-contract.md) and link to it one level deep.

Trim narrative asides (e.g. "The cost is real but the discipline buys auditability", "routinely saves a full re-run") that do not change the executable guidance.

Dimension	Reasoning	Score
Conciseness	Dense and operational with no padding explaining basic concepts, but the manager/worker model and the "don't use --full-auto" warning are restated across sections and a few narrative asides could be trimmed, keeping it just below the lean score-5 anchor.	4 / 5
Actionability	Provides fully executable guidance: a concrete codex exec invocation with real flags, PID-wait monitoring, git status/diff recovery, before/after HEAD integrity checks, and an explicit JSON output contract, covering the common dispatch cases.	5 / 5
Workflow Clarity	Multi-step processes are clearly sequenced with explicit validation checkpoints and a feedback loop (reviewer boundary violation -> reset/clean -> re-dispatch -> escalate after two failures), matching the score-5 anchor including error recovery for the destructive corrector path.	5 / 5
Progressive Disclosure	Well-organized with clear section headers and a verified one-level-deep reference to scripts/detect_sandbox_spiral.sh, but the long inline reviewer JSON contract and orchestrator-side enforcement block are bulk that could live in a separate reference file, keeping it just below the score-5 anchor.	4 / 5
	Total	18 / 20 Passed

Description

95%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, well-targeted description: it states the capability concisely, supplies natural trigger phrases, answers both what and when, and explicitly disambiguates from a related skill. The only minor weakness is that capabilities are listed as topic areas rather than as discrete concrete verbs.

Dimension	Reasoning	Score
Specificity	Names the codex exec domain and several concrete coverage areas (flags, sandbox traps, monitoring, recovery), but frames them as topic coverage rather than discrete concrete actions, keeping it just below the comprehensive score-5 anchor.	4 / 5
Completeness	Explicitly answers both what (delegate a single coding task to codex exec; covers flags, sandbox traps, monitoring, recovery) and when ("Use when..." with concrete trigger phrases and a negative boundary), matching the score-5 anchor.	5 / 5
Trigger Term Quality	Provides multiple natural user phrases ("hand off to codex", "run codex on this", "dispatch codex on this ticket") plus the "one-shot invocation" synonym, giving comprehensive coverage of terms users would actually say.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (single-shot codex exec) with distinct triggers and an explicit disambiguation against the sibling codex-issue-waves skill, minimizing conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: ddnetters/homelab-agent-skills
Path: invoking-codex-exec/SKILL.md
Commit: 808c382

Reviewed: 8 days 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.