CC odl-review
Decide whether an opendataloader-pdf pull request is safe to merge, by its results, inside clear bounds: the change does nothing harmful or hidden; it fixes its target (what the issue or the pull request asks), shown with the code, tests, and runs; and nothing that works on main breaks. If these hold, approve; do not ask for more. Code enforces what must not vary (the trust boundary of the trees, the scan, the earlier decisions, the test-suite comparison, and the verdict itself), you judge the target, the risks, and the checks, and the user approves what cannot be undone. You doubt your plan, the change, and the evidence once. Other people's code runs only in an isolated environment. Use it when the user asks "review #123", "can I merge #123", "is this PR safe", or wants a merge verdict. Also use it to triage all open pull requests ("which PRs should I look at first"). Do not use it to fix an issue. The user approves twice: the plan at the start, and the answers and posts at the end.
Decide whether an opendataloader-pdf pull request is safe to merge, by its results, inside clear bounds: the change does nothing harmful or hidden; it fixes…
As a process C 56/100 · Has gaps — References files that are not bundled: scripts/pr-reminders.py
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 · 1
✓ No critical or high findings
Medium and low: 1
-
low Risky intent
intent-offensive-securitySKILL.md:134Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| I4 | No privilege escalation | Limits an escape |
Files scanned: 21. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 12498 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: scripts/pr-reminders.py
Process rating: all ten parameters 56/100
- 0Tools and files. 1 referenced file(s) missing: scripts/pr-reminders.py
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 30 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 40Execution cost. Instruction body is 12498 tokens: crowds the task out of the window
- 100Steps. 88 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 9 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- low 19 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (21 tags): a typed call is more reliable
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
- +3Description length 998: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -2localhost URLs: will not work for another user
- -34 of 13 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +5Description quotes 4 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 20 headings
- +3Step-by-step instructions: 88 items
- +4Has examples (8 code blocks)
- +4Reference files are cited in the instructions (5 of 5)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 74.