CD paper-review-lite
Provides a pre-submission manuscript audit covering argument, numerical consistency, references and DOIs, writing, figures, CONSORT flow, pre-registration, and the replication archive. Use before paper submission or when a draft needs an adversarial, quote-grounded review with a journal-readiness checklist. Add `--codex` to run the specification independently on Claude and Codex, GPT-6 Astra, and cross-check findings.
Provides a pre-submission manuscript audit covering argument, numerical consistency, references and DOIs, writing, figures, CONSORT flow, pre-registration…
As a process D 45/100 · Unfinished process — References files that are not bundled: ../orchestrate/SKILL.md
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- The text references files that are not there: add them or drop the references.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 6
✓ No critical or high findings
Medium and low: 6
-
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: AskUserQuestion Read Write Edit Bash Glob Grep WebSearch WebFetch Agent
-
low Risky intent
intent-offensive-securitySKILL.md:38Offensive-security / dual-use content (legitimate for authorised testing; review intended use)**Sub-agent model and effort assignment.** Route by task type, not by agent number — argument-level dimensions (evaluating logical structure, cross-referencing claims, detecting silent discrepancies)
-
low Risky intent
intent-offensive-securitySKILL.md:74Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)**Agent 1 — Content & Argument (Red Team primary)**: Read the full paper. Your posture is adversarial but fair: find every place the argument is weaker than the paper presents it to be. Check logical
detector -
low Risky intent
intent-offensive-securitySKILL.md:188Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)Two reviewers — Claude (the orchestrator) and Codex (GPT-6 Astra at `xhigh` reasoning effort by default, invoked via Bash `codex exec`) — independently apply the § 2 specification to the same paper, t
quoted -
low Risky intent
intent-offensive-securitySKILL.md:190Offensive-security / dual-use content (legitimate for authorised testing; review intended use)Roughly 22 model calls total (9 Claude Red Team, 9 Codex Red Team, 4 cross-model Blue Team) plus orientation and synthesis by the orchestrator. Reach for it before submission when you want maximum adv
-
low Risky intent
intent-offensive-securitySKILL.md:192Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)**`--codex-effort <low|medium|high|xhigh>`** sets the reasoning effort for every Codex call in the run (default `xhigh`). Use it when the paper is short, the run is a mid-draft pass rather than the fi
quoted
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 11546 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: ../orchestrate/SKILL.md
Process rating: all ten parameters 45/100
- 0Tools and files. 1 referenced file(s) missing: ../orchestrate/SKILL.md
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 40Execution cost. Instruction body is 11546 tokens: crowds the task out of the window
- 50Failures and branches. 0 branches, has a failure section
- 60Result and completion. Output format stated, no completion criterion
- 85Steps. 54 steps, 1 vague phrases
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +1No license
- +2Single-language instructions
- +3Description length 421: enough signal without eating the budget
- +4Structure: 17 headings
- +3Step-by-step instructions: 54 items
- +3Output format is stated explicitly
- +4Has examples (6 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.