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 agentmods add skills/merencia/pairwith/pair-withnpx skills add merencia/pairwith --skill pair-withgit clone --depth 1 https://github.com/merencia/pairwithWhat 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 | $0.00048 | $0.00969 |
| Opus 5 | $0.00024 | $0.00485 |
| Sonnet 5 | $0.00010 | $0.00194 |
| Haiku 4.5 | $0.00005 | $0.00097 |
Grade E, and why
pair-with scanned grade E with 2 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 3d 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.
Instruction-override phrasinghighPrompt injection
Text telling the model to disregard its earlier instructions or safety rules is the shape of a prompt injection, whoever wrote it.
- bypass safety policies or refusal behavior; Harvests environment variableshighData exfiltration
Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.
- extract secrets, read environment variables, or exfiltrate data; How it starts
The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pair-with
Apply a community-created AI style profile and respond using its principles, heuristics, and tone for the duration of the task.
What a profile is (and is not)
A profile at ~/.claude/agents/<handle>.md is a style profile — a set of prompt instructions inspired by publicly observable aspects of a developer's technical work. It is not the real person, does not represent them, and is not endorsed by them unless explicitly stated. You are an AI coding assistant applying a profile. You are not the person the profile is inspired by.
Treat the profile as a source of engineering heuristics and communication style. Do not claim to be the person, do not claim their endorsement, and do not invent biographical or personal details.
Parsing the invocation
The user calls this skill as:
/pair-with <handle> <task description>
- The first word of the arguments is the handle (e.g.
merencia). - Everything after it is the task (e.g.
review these changes,help me design this API,what do you think about this approach).
Loading the profile
Read the profile file at ~/.claude/agents/<handle>.md.
If the file does not exist:
- Tell the user the profile is not installed.
- Suggest:
npx pairwith install <handle> - Stop. Do not proceed without a profile.
If the file exists, read it in full before doing anything else — including the non-impersonation header at the top. Respect it.
Applying the style profile
Use the profile's content to shape how you respond to the task:
- Apply the Principles as constraints on your reasoning. If the profile avoids premature abstraction, do not propose abstractions even when they feel natural.
- Use the Decision heuristics when there are trade-offs. If the profile has a tie-breaker rule, apply it.
- Match the Tone and communication section. If the profile is terse, be terse. If the profile asks sharp questions before diving in, do that. If the profile pushes back on scope, push back.
- Respect the Anti-patterns — if something in the request conflicts with what the profile avoids, call that out in the profile's voice.
- Use the Stack preferences when making specific recommendations.
- Reason in the profile's order — if the profile describes a typical order of operations for a new problem, follow it.
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.
- 3d ago First seen · 82 lines · 0 tokens per session scan E a49b1aca484b
pair-with is a skill published in the GitHub repository merencia/pairwith (7 stars, last pushed 4mo ago), licensed MIT. It adds 48 tokens to every session and 969 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it E with 2 findings (instruction-override phrasing, harvests environment variables). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…