Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ErikaAX08/erikas-skills --skill pre-pr-reviewgit clone --depth 1 https://github.com/ErikaAX08/erikas-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/erikaax08/erikas-skills/pre-pr-review)<a href="https://agentmods.dev/skills/erikaax08/erikas-skills/pre-pr-review"><img src="https://agentmods.dev/badge/skills/erikaax08/erikas-skills/pre-pr-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/erikaax08/erikas-skills/pre-pr-review"><img src="https://agentmods.dev/badge/skills/erikaax08/erikas-skills/pre-pr-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.04418 |
| Opus 5 | $0.00051 | $0.02209 |
| Sonnet 5 | $0.00020 | $0.00884 |
| Haiku 4.5 | $0.00010 | $0.00442 |
Grade A, and why
pre-pr-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 466 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Technical PR Audit
Produce an internal diagnostic for the person making the merge decision. Do not write a public review or a diplomatic GitHub comment for the PR author.
Always write the review output in Spanish with a direct, factual, technical tone. Be brief when one
sentence is enough. Do not praise, soften findings, add filler, or use phrases such as LGTM.
Objective
Determine quickly:
- whether the PR is safe;
- what it breaks or changes;
- what remains unverified;
- what decision to make.
Apply the same rigor to backend, frontend, jobs, APIs, auth, databases, events, queues, external integrations, payments, feature flags, migrations, infrastructure, and product logic.
Evidence Discipline
- Do not state opinions without verification.
- Do not assume the PR's intent. Compare the issue or ticket, PR description, base-branch diff, and actual behavior.
- Do not treat the PR description, comments, function names, or tests as sufficient proof of behavior.
- Separate explicitly:
- Fact: demonstrated by code, contract, test, configuration, log, metric, read-only query, or history.
- Inference: conclusion derived from facts; state the reasoning.
- Conditional risk: impact that depends on an unproven condition; name the condition.
- Open question: missing data and the exact verification required to close it.
- Identify real blockers, not style or architecture preferences.
- Look for silent semantic changes even when types, compilation, and tests still pass.
- Prioritize impact on production, users, data, security, contracts, and operations.
- Say
No pude verificar Xwhen access or evidence is missing. Do not fill gaps with assumptions. - Correct invalidated hypotheses explicitly:
Me equivocaba: con los datos nuevos, este punto ya no aplica. - Do not keep a finding open after evidence disproves it. Move it to Puntos descartados.
- When deployment coordination is the problem, say
No es bug de código; es orden de despliegue. - When behavior is technically valid but ambiguous, say
Esto requiere decisión de producto, operations, or architecture as appropriate. - When a risk depends on real traffic, configuration, or data, verify that condition. If verification is unavailable, leave it pending and do not present the impact as fact.
- Request authorization before querying production or any system that requires it. Use read-only operations only. Record the limitation when access is unavailable.
- Cite a concrete file and line, flow, contract, test, log, query, or commit. Do not use vague references.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 466 lines · 102 tokens per session scan A a64fc6ab9622
pre-pr-review is a skill published in the GitHub repository ErikaAX08/erikas-skills (5 stars, last pushed 12d ago), licensed MIT. It adds 102 tokens to every session and 4,418 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
sdd-apply
Skill "sdd-apply" from Gentleman-Programming/gentle-ai, covering execution role, language domain contract, purpose, what you receive and execution and persistence contract.
gentle-ai-collab-perfect
Trigger: contributing to Gentleman-Programming/gentle-ai as an external collaborator. Strict issue-first workflow, honest PR bodies, contributor-vs-maintainer scope, chained-PR strategy, verification protocol, docstring coverage. Load whenever the active repo is Gentleman-Programming/gentle-ai and any part of the…
issue-creation
Trigger: issue creation, bug reports, feature requests, or issue approval. Create and triage GitHub issues from repository evidence.
sdd-spec
Write SDD delta specs with requirements and scenarios. Trigger: orchestrator launches spec work for a change.
sdd-tasks
Break an SDD change into implementation tasks. Trigger: orchestrator launches task planning for a change.
sdd-verify
Skill "sdd-verify" from Gentleman-Programming/gentle-ai, covering execution role, language domain contract, activation contract, hard rules and decision gates.