An evidence-based investigation method for finding root causes, bugs, inconsistencies, and hidden problems in code, systems, performance, architecture, or data. It examines the evidence through deduction, user intent, and a search for missing cases.

Use it to debug or investigate behavior, trace evidence through a system, uncover overlooked edge cases, and explain the likely cause for someone else to fix.

It turns a confusing failure into a documented case by testing assumptions and tracing what the system actually does. It reports findings without changing the code.

investigate — Skill for Claude Code by ondrej-svec

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code

/plugin marketplace add ondrej-svec/heart-of-gold-toolkit

Claude Code

/plugin install deep-thought

Made for: Claude Code.

Or install deep-thought, the plugin that ships this one along with the rest of its 11 skills, 9 agents.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/investigate/github.svg)](https://agentmods.dev/skills/ondrej-svec/heart-of-gold-toolkit/investigate)

Your own site

<a href="https://agentmods.dev/skills/ondrej-svec/heart-of-gold-toolkit/investigate"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/investigate/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/ondrej-svec/heart-of-gold-toolkit/investigate"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/investigate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 92 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,604 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00092	$0.02604
Opus 5.5	$0.00037	$0.01042
Sonnet 5.5	$0.00018	$0.00521
Haiku 4.5	$0.00009	$0.00260

Measured 2mo ago against content hash 9205b57e75ed, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

investigate 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 2mo 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.

**This skill MAY NOT:** edit code, fix issues, create PRs, deploy. The only Bash allowed is read-only diagnostics (git log, curl for status, kubectl get, etc.).

plugins/deep-thought/skills/investigate/SKILL.md · 285 lines

How it starts

The opening of the file, as written. The whole thing — 285 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Investigate

Detective-style investigation. Follow evidence trails, notice what's missing, connect what others overlook.

Boundaries

This skill MAY: read code, run diagnostic commands (read-only), trace evidence, present findings. This skill MAY NOT: edit code, fix issues, create PRs, deploy. The only Bash allowed is read-only diagnostics (git log, curl for status, kubectl get, etc.).

This is an investigation, not a fix. Present the case — the user decides what to do.

The Three Minds

Three investigative lenses — reasoning frameworks that each unlock a different class of problems:

Sherlock Holmes — Deductive elimination. What MUST be true? Verify each premise. Catches: logic errors, broken invariants, impossible states, type mismatches.

Hercule Poirot — Psychological method. Study intent and mental model. The gap between what someone THOUGHT the system does and what it ACTUALLY does. Catches: misunderstood APIs, wrong assumptions, subtle misreads.

Columbo — Persistent nagging. Something doesn't sit right. "Just one more thing..." Catches what's MISSING — error handlers, edge cases, tests, cleanup.

Common Rationalizations

Shortcut	Why It Fails	The Cost
"Jump to Phase 5 — I already know the answer"	Skipping evidence gathering confirms biases, not bugs	Wrong diagnosis → wrong fix → problem persists
"This code looks fine, move on"	Dangerous bugs LOOK correct	The bug you didn't investigate ships
"Not my scope — skip it"	Evidence trails cross boundaries	Surface symptom fixed, root cause remains
"The tests pass, so it's correct"	Tests test what the author THOUGHT, not what it ACTUALLY does	False confidence

Phase 0: Establish the Case

Entry: User described a problem, pointed at code, or said "something's wrong."

Auto-detect polarity from context:

Signals	Polarity	Focus
Diff, file path, error in code	Code	Bugs, logic errors, type mismatches
"Slow", latency, timeouts	Performance	Queries, memory, bottlenecks
"Is this the right structure"	Architecture	Patterns, coupling, design drift
"Data doesn't look right"	Data	Integrity, schema, migration risks
"Something is broken in prod"	System	Infrastructure, networking, runtime

Read the full file on GitHub · 285 lines

Changes

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.

2mo ago First seen · 285 lines · 92 tokens per session scan A 9205b57e75ed

Subscribe to this mod's changes

investigate is a skill published in the GitHub repository ondrej-svec/heart-of-gold-toolkit (19 stars, last pushed 1mo ago), licensed MIT. It adds 92 tokens to every session and 2,604 once invoked, about $0.0004 per session on Opus 5.5. 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.

Debugging

investigate next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
investigateondrej-svec/heart-of-gold-toolkit	Skill	19	1mo ago	MIT	A	92	Claude Code
systematic-debuggingopen-metadata/OpenMetadata	Skill	16k	today	Apache-2.0	A	37	Claude Code
log-error-digestzebbern/claude-code-guide	Skill	4.7k	today	MIT	A	71	Claude Code, Codex
cwv-optimizeradobe/skills	Skill	319	today	Apache-2.0	A	112	Claude Code
ci-fixersipyourdrink-ltd/bernstein	Skill	1.2k	yesterday	Apache-2.0	A	15	Claude Code, Codex
diagnosing-missing-recordingsPostHog/posthog	Skill	41k	today	none	A	76	Claude Code, Codex
loop-engineeringhuytieu/COG-second-brain	Skill	1.3k	15d ago	MIT	A	63	Claude Code

investigate