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/jeonje/claude-plugins/hypothesis-debugnpx skills add JeonJe/claude-plugins --skill hypothesis-debuggit clone --depth 1 https://github.com/JeonJe/claude-pluginsWrote 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/jeonje/claude-plugins/hypothesis-debug)<a href="https://agentmods.dev/skills/jeonje/claude-plugins/hypothesis-debug"><img src="https://agentmods.dev/badge/skills/jeonje/claude-plugins/hypothesis-debug.svg" alt="Measured on agentmods" 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 | $0.00107 | $0.00861 |
| Opus 5 | $0.00053 | $0.00430 |
| Sonnet 5 | $0.00021 | $0.00172 |
| Haiku 4.5 | $0.00011 | $0.00086 |
Grade A, and why
hypothesis-debug 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 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.
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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hypothesis-driven debugging
Force a systematic path from symptom to root cause: facts-only intake → parallel evidence collection → competing hypotheses (≥2) with mandatory falsification → discriminating probe → report with regression prevention. Never "just try a fix" on a guess.
References index
| File | When to read |
|---|---|
| references/hypothesis-lanes.md | Phase 3 — building the competing-hypothesis set; Phase 4+ — 5-why vertical descent |
| references/troubleshoot-report-template.md | Phase 5 — report skeleton |
Phases
1. Intake — facts only
Record: symptom in one line; when it started; affected scope (users/requests/time window); what changed recently (deploys, config, data); and explicitly, what is NOT yet known. No hypotheses at this phase — an early guess anchors the whole investigation.
2. Parallel evidence collection
Collect from every available source in parallel, read-only: application/error logs, DB state (read-only queries), metrics/APM, reproduction attempt in a safe environment. Label each finding with its source and timestamp. Time-zone discipline: convert all timestamps to one zone before correlating — mixed-zone timelines fabricate causality.
3. Localize + hypothesize (competing, ≥2)
Read hypothesis-lanes.md. Draw at least one hypothesis from each of the three lanes — code logic / environment & data / measurement error — so a single fixation cannot capture the investigation. For each hypothesis maintain:
| Hypothesis | Evidence for | Evidence against | Discriminating probe | Status |
|---|
Mandatory falsification: for the leading hypothesis, actively look for evidence that would disprove it before promoting it. A hypothesis you never tried to kill is not confirmed.
4. Probe loop
Design the cheapest probe that discriminates between the top hypotheses (a log line, a targeted query, a minimal reproduction). Run it, then promote or falsify. Repeat until one hypothesis survives with direct evidence. If the same probe fails twice, stop gathering more of the same logs — restate hypothesis/observation/next-probe in three lines and pick a different check.
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.
- 3d ago First seen · 56 lines · 107 tokens per session scan A f34166a606f9
hypothesis-debug is a skill published in the GitHub repository JeonJe/claude-plugins (3 stars, last pushed 29d ago), licensed MIT. It adds 107 tokens to every session and 861 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
api-testing
HTTP API testing with Supertest (TS) and httpx/pytest (Python). Use when the user mentions API testing, Supertest, httpx, REST/GraphQL validation, or HTTP response errors.
researching-code
Use this skill when the user asks an architectural or semantic question about a codebase — phrases like "how does X work?", "what's the architecture?", "help me understand this codebase", "find all components that use Y", "trace the data flow from A to B", "where is feature Z handled", "I'm new to this code, where do…
dev-environment-bootstrapping
Use this skill when the user asks to bootstrap, set up, create, or initialize a Shopware development environment from scratch — phrases like "set up a Shopware dev environment", "clone and install Shopware", "initialize a Shopware plugin project", "bootstrap Shopware and a new plugin called X", "get a fresh Shopware…
phpunit-integration-to-unit-migrating
Use this skill ONLY when the user explicitly requests an audit, migration, or evaluation of whether a Shopware integration test belongs in the unit suite — trigger phrases like "audit integration tests", "migrate integration tests to unit", "is this an integration test or a unit test", "evaluate integration tests for…
cross-browser-typography-qa
Use this skill when web text looks clipped, flattened, wrapped incorrectly, misaligned, or materially different between Chromium and WebKit/Safari. Treat text rendering as both a geometry problem and a visual paint problem: passing DOM bounds does not prove that every glyph is intact.
ryk
Use when Pi tool calls are protected by ryk runtime guardrails or when a secret was captured from a Pi prompt.