codex-huge-context

Codex 1M context: direct OpenAI Responses API inference, safe Sol/Terra/Luna input headroom, Keychain delivery, and Mac fleet rollout.

Quality

71%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/codex-huge-context/SKILL.md

Quality

Content

92%

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

The content is highly actionable and lean, with concrete config values, executable commands, clear sequenced workflows, and explicit validation gates backed by a real preflight script. Its main weakness is progressive disclosure: dense operational detail is kept inline rather than split into well-signaled one-level-deep reference files.

Suggestions

Extract the long TOML/JSON/zsh configuration blocks and the per-host fleet-checklist detail into one-level-deep reference files (e.g. reference the preflight script path more prominently and move catalogue/provider config to a CONFIG.md), keeping SKILL.md as a concise overview.

Consider surfacing the bundled preflight.test.rb as part of the verification section so the validation tooling is discoverable alongside preflight.rb.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence throughout, with no padded explanations of what Codex, OAuth, or context windows are; nearly every line carries a specific number, command, or value, matching the 'lean and efficient; every token earns its place' anchor.	3 / 3
Actionability	It provides fully executable, copy-paste-ready guidance — exact JSON catalogue values, complete TOML provider blocks, a zsh Keychain helper script, and concrete verification commands with expected outputs ('922000', '922000', '700000') — matching the 'fully executable code/commands; copy-paste ready' anchor.	3 / 3
Workflow Clarity	Multi-step processes are clearly sequenced with explicit validation checkpoints, notably the preflight gate ('Do not mark a rollout complete or launch Codex when this fails'), the ordered fresh/resumed-session restart steps, and a Failure Policy with validate->fix->retry feedback loops for risky batch and remote operations.	3 / 3
Progressive Disclosure	Sections are well-organized and the one bundle reference (scripts/preflight.rb, which exists alongside preflight.test.rb) is correctly signaled, but the body is a fairly monolithic ~157-line inline wall of TOML/JSON/zsh blocks and operational detail rather than an overview pointing to one-level-deep reference docs, so content that could be split out remains inline.	2 / 3
	Total	11 / 12 Passed

Description

50%

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 to a clear niche but reads as a compressed keyword list of capabilities rather than a trigger-oriented sentence; it lacks an explicit 'Use when...' clause and leans heavily on internal jargon (Sol/Terra/Luna, input headroom).

Suggestions

Add an explicit trigger clause such as 'Use when configuring, repairing, or auditing Codex's 1M context setup' to satisfy the 'when should Claude use it' requirement.

Reword in third-person action voice naming concrete actions (e.g. 'Configures, repairs, and audits Codex's 1M-token context setup') instead of a noun-phrase list to raise specificity.

Soften jargon or pair it with natural terms users would actually say (e.g. '1M context window', 'long-context Codex') so the description matches real requests more reliably.

Dimension	Reasoning	Score
Specificity	Names a concrete domain and several sub-capabilities ('direct OpenAI Responses API inference, safe Sol/Terra/Luna input headroom, Keychain delivery, and Mac fleet rollout'), but these are topic noun-phrases rather than a list of multiple concrete actions Claude performs, so it does not reach the 'lists multiple specific concrete actions' anchor.	2 / 3
Completeness	It answers 'what does this do' with a capability list, but there is no 'Use when...' clause or equivalent explicit trigger guidance for when Claude should use it, which per the judging guidelines caps completeness at 2.	2 / 3
Trigger Term Quality	It includes some natural terms a user might say ('Codex 1M context', 'Mac fleet rollout'), but the phrase is dominated by technical jargon ('Sol/Terra/Luna input headroom', 'Responses API inference') and is missing common natural variations, so coverage is partial rather than strong.	2 / 3
Distinctiveness Conflict Risk	It describes a distinct niche (Codex 1M context configuration) unlikely to fire for unrelated skills, but the jargon-heavy phrasing makes the triggers less crisp than the 'clear niche with distinct triggers' anchor, so it lands at 'somewhat specific but could still overlap' rather than fully distinct.	2 / 3
	Total	8 / 12 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: steipete/agent-scripts
Path: skills/codex-huge-context/SKILL.md
Commit: bb36883

Reviewed: about 13 hours 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.