coding-investigation

A read-only procedure for investigating unfamiliar code behavior or tracing a specific symbol, such as a function or class. It follows the symbol’s activation, callers, data flow, configuration, and runtime connections using source evidence.

In plain words
What is it for?
Finding where a behavior or symbol is defined, what calls it, what it calls, how it is configured or registered, and which repositories or files it affects.
Why use it?
Unfamiliar code can make it difficult to identify what starts a behavior, which code owns it, or what it affects. This procedure limits the investigation to a defined question and avoids changing files.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/evoelsewhere/evoflux/coding-investigation
Any agent
npx skills add evoelsewhere/evoflux --skill coding-investigation
Clone the repo
git clone --depth 1 https://github.com/evoelsewhere/evoflux

Made for: Claude Code, Codex.

Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,274 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00081 $0.01274
Opus 5 $0.00041 $0.00637
Sonnet 5 $0.00016 $0.00255
Haiku 4.5 $0.00008 $0.00127

Measured 2d ago against content hash 8e3dec3ee3a2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coding-investigation 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 2d 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.

app/agent/builtin_skills/coding-investigation/SKILL.md · 131 lines

How it starts

The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Investigate code behavior

Remain read-only. Prove only the requested entry point, deciding condition, and downstream effect, then stop. Do not load bundled references unless a condition under Gaps requires one.

Choose one lane

  • Exact symbol: the request or observed source names a function, method, class, constant, or qualified symbol. Skip code_context with action="search" and call native code_context immediately.
  • Unknown root: the request names behavior, UI text, route, field, tag, configuration key, event, or runtime effect. Run one code_context with action="search" using that stable artifact, select one declared identifier, then promote it to graph.

In a multi-repository workspace, omit repository, paths, and languages from that first unknown-root search unless the user names the target or direct source evidence already identifies it. The primary-workspace marker and a repository's language are hints, not proof that it owns the requested behavior. When a hit belongs to a non-primary repository, keep that repository identity: ordinary filesystem tools need the absolute path shown in the repository map. Do not strip it to a primary-relative path.

A single bounded investigation question is one work lane, not an implementation plan. Do not create a todo unless the request contains multiple independent questions or deliverables. When several independent source ranges are already known, issue their read-only calls together in one model turn instead of serializing them.

read results already carry stable line numbers. Reuse those anchors for the final answer; never rerun sed, nl, or another read solely to add citation numbers that the existing result already provides.

Never pass request prose, a filename, module, route, comment, or guessed spelling as the exact-symbol query. Promotion from unknown-root to exact-symbol ends broad discovery.

exact declared symbol known
  -> code_context(action=smallest operation, query=symbol)

unknown root
  -> code_context(action="search", query=observed artifact, refresh=true)
  -> code_context(action=smallest operation, query=one returned declared identifier)
  -> targeted source only for one named semantic gap
  -> answer

Read the full file on GitHub · 131 lines

Files

What ships with it

4 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.

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.

  1. 2d ago First seen · 131 lines · 81 tokens per session scan A 8e3dec3ee3a2

Subscribe to this mod's changes

coding-investigation is a skill published in the GitHub repository evoelsewhere/evoflux (5 stars, last pushed 5d ago), licensed Apache-2.0. It adds 81 tokens to every session and 1,274 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

kane-cli

Browser automation + AI test authoring via kane-cli - run browser objectives, generate & refine test scenarios/cases from a description, design requirement-linked test suites from a PRD/spec (assurance), parse NDJSON output, inspect logs, save runnable test.md. Use for any task requiring a real browser (navigate…

LambdaTest/kane-cli · 159 tokens

control-chrome

Control the user's Chrome browser for tasks that depend on existing Chrome state: tabs, logged-in sessions, or extensions. Prefer purpose-built connectors, APIs, or CLIs when available.

rojim666/SztuCode · 40 tokens

recent-release

Render Rove's recent releases as a self-contained, Rove-styled HTML page (Chinese by default). Reads shipped notes from packages/kobe/CHANGELOG.md + git tags, filters to user-facing changes, and emits one standalone .html file in the claude theme (terracotta on near-black, JetBrains Mono, terminal chrome). Use when…

Sma1lboy/rove · 112 tokens

yutori-scout

Set up continuous web monitoring with Yutori Scouts. Use when the user wants to track news, competitors, product updates, funding rounds, price changes, or any recurring web information.

yutori-ai/yutori-mcp · 43 tokens

plan-review-triggers

Required reading whenever drafting any multi-step proposal, implementation plan, refactor approach, or response containing file paths and decision points. Use BEFORE writing any response that lists 3+ file paths, 2+ option-tradeoffs, a "Plan:"/"Steps:"/"Concrete changes:" section, or ends with "OK to proceed?" /…

ndom91/open-plan-annotator · 199 tokens

yutori-computer-use

Run local Mac desktop tasks with Yutori computer use. Use when the user wants to operate macOS apps, websites in a local browser, or cross-app workflows on the visible desktop.

yutori-ai/yutori-mcp · 45 tokens