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 joneslloyd/agent-skills --skill counterpart-analysisgit clone --depth 1 https://github.com/joneslloyd/agent-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/joneslloyd/agent-skills/counterpart-analysis)<a href="https://agentmods.dev/skills/joneslloyd/agent-skills/counterpart-analysis"><img src="https://agentmods.dev/badge/skills/joneslloyd/agent-skills/counterpart-analysis/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/joneslloyd/agent-skills/counterpart-analysis"><img src="https://agentmods.dev/badge/skills/joneslloyd/agent-skills/counterpart-analysis.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 | $0.00183 | $0.06218 |
| Opus 5 | $0.00092 | $0.03109 |
| Sonnet 5 | $0.00037 | $0.01244 |
| Haiku 4.5 | $0.00018 | $0.00622 |
Grade A, and why
counterpart-analysis 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 9d 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 — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Counterpart Analysis
Read a conversation the way a skilled negotiator or seasoned operator would: separate what people say from what they want, identify the constraints they're operating under, and surface the phrasing-level evidence — while staying honest about confidence, labeling every input by its epistemic status, and always holding a competing explanation.
The output is never "here is what this person is thinking" (unknowable). It is "here are 2–3 ranked hypotheses about their angle, here is the specific evidence for each, here is which parts rest on fact versus assumption, here is what would disconfirm each, and here is what to ask or watch for next to find out."
Core stance
Four principles govern everything in this skill:
-
Interests, not positions. What someone asks for is a position ("we need this by Q3"). What drives the ask is an interest (their bonus depends on a Q3 launch; their boss is watching this project; they've been burned by slipping vendors). Positions are visible in the text; interests must be inferred. Always work one level below the stated position, and say explicitly when you are inferring.
-
Evidence before interpretation. Every claim about someone's motivation must be anchored to a specific quote, phrasing choice, omission, behavioral moment, or piece of known context. If you cannot point to it, mark it as an assumption. Format: quote → observation → plausible reading(s) → confidence.
-
Provenance on everything. Every input to a conclusion carries one of four tags (defined below):
[OBSERVED],[CONTEXT],[INFERRED],[ASSUMED]. The reader must be able to look at any hypothesis and see immediately which parts rest on the material, which on supplied background, which on reasoning, and which on unverified assumption — and what breaks if an assumption is wrong. -
Competing hypotheses, always. The failure mode of this kind of analysis is confident overreading — a paranoid narrative built on ambiguous phrasing. For the primary "angle" hypothesis, always construct at least one rival explanation (usually the innocuous one: they're busy, they're a non-native speaker, that's just corporate register, they genuinely mean it). Rank hypotheses and state what evidence would separate them. If the innocuous explanation fits the evidence equally well, say so prominently.
What ships with it
2 files 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.
- 9d ago First seen · 213 lines · 183 tokens per session scan A 30bc8fa180ea
counterpart-analysis is a skill published in the GitHub repository joneslloyd/agent-skills (2 stars, last pushed 2mo ago), licensed MIT. It adds 183 tokens to every session and 6,218 once invoked, about $0.0009 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…