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 arpitbbhayani/ape-skills --skill ape-poke-holesgit clone --depth 1 https://github.com/arpitbbhayani/ape-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/arpitbbhayani/ape-skills/ape-poke-holes)<a href="https://agentmods.dev/skills/arpitbbhayani/ape-skills/ape-poke-holes"><img src="https://agentmods.dev/badge/skills/arpitbbhayani/ape-skills/ape-poke-holes/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/arpitbbhayani/ape-skills/ape-poke-holes"><img src="https://agentmods.dev/badge/skills/arpitbbhayani/ape-skills/ape-poke-holes.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00118 | $0.03984 |
| Opus 5 | $0.00059 | $0.01992 |
| Sonnet 5 | $0.00024 | $0.00797 |
| Haiku 4.5 | $0.00012 | $0.00398 |
Grade A, and why
ape-poke-holes scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
2. **A URL**: a string starting with `http://` or `https://`. Use `WebFetch` to download the content, then strip navigation, sidebars, footers, and comments before working on the body. If the URL points to a PDF, downloa How it starts
The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Poke Holes Skill
An adversarial reviewer. Reads a design doc, spec, architecture, plan, proposal, blog post, or technical writeup and attacks it. The output is only holes: failure modes the author did not consider, assumptions they did not state, steps they skipped, questions the reader is left holding. No praise, no balance, no "overall this looks solid". If a section survives the attack, it simply does not appear in the output.
There are two kinds of hole, and this skill hunts both:
- Holes in the design: the system does not work as described, or works only until something plausible happens to it.
- Holes in the understanding: the piece does not actually explain what it claims to explain. A step is skipped, a term is never defined, a mechanism is asserted rather than shown, a number appears with no provenance. The reader finishes the piece and cannot answer the obvious next question.
The second kind matters as much as the first. A design nobody can reconstruct from the doc is a broken design. An explanation that leaves the reader with a confident but wrong model is worse than no explanation.
The skill pokes holes; it does not fill them. Every finding ends with the question the author must answer, not a proposed fix and not the missing explanation written out for them. Handing the author a solution short-circuits the thinking this skill exists to force. The author closes the holes; the ape only finds them.
Scope: this skill attacks substance, not style. Clunky sentences, weak headings, passive voice, and paragraph order are editing concerns and out of scope -- ape-cut-fluff and ape-review-blog own those. Wording becomes a finding only when it costs comprehension or precision: two engineers would build two different systems from it, or a reader would walk away believing something false.
What Counts as a Hole
Attack the document along these lines. Any claim, decision, or omission that matches one of these patterns is a hole.
Holes in the design
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 · 182 lines · 118 tokens per session scan A 8d6094f36301
ape-poke-holes is a skill published in the GitHub repository arpitbbhayani/ape-skills (40 stars, last pushed yesterday), licensed MIT. It adds 118 tokens to every session and 3,984 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
best-practices
Apply modern web development best practices for security, compatibility, and code quality. Use when asked to "apply best practices", "security audit", "modernize code", "code quality review", or "check for vulnerabilities".
resolve-pr-comments
Evaluate, fix, answer, and reply to GitHub pull request review comments and conversation comments. Handles both change requests (fix or skip) and reviewer questions (explain using reasoning recalled from past Claude Code transcripts). Use when the user asks to "resolve PR comments", "fix review comments", "address PR…
review-code
Review code for bugs, security vulnerabilities, API misuse, consistency issues, simplicity problems, or test coverage gaps by running internal reviews and a peer review in parallel and returning combined findings. Single-concern with a type argument, or full review with no argument. Use when the user asks to "review…
assess-technical-debt
Assess project-wide structural technical debt: complexity hotspots, deprecated API usage, duplication clusters, and architecture rot. Ranks findings by impact and refactor effort into a report at .turbo/technical-debt.md. Use when the user asks to "assess technical debt", "find technical debt", "review technical…
map-codebase
Deep architecture report that fans out parallel inspections across different aspects of the codebase (structure, tech stack, APIs, patterns, data flow, dependencies, testing) and synthesizes findings into a comprehensive document at .turbo/codebase-map.md and .turbo/codebase-map.html. Use when the user asks to "map…
peer-review
Run an independent peer review via Codex. Use when the user asks to "peer review", "peer review my code", "peer review my plan", "get a second opinion", or "independent review".