investigate SKILL.md skillA

investigate from gabrielbsandim/task-funnel is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add gabrielbsandim/task-funnel/investigate@git:20261020.5ec3e90

Pins git:20261020.5ec3e90 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
gabrielbsandim/task-funnel
Path
skills/investigate/SKILL.md
Versions recorded
2
Latest version
git:20261020.5ec3e90, committed Oct 20, 2026
First version recorded
Aug 17, 2026
Size
2,770 bytes
Repository stars
3
License
MIT

Investigation flow. Takes a question about the system (why does X happen, where does Y live, is Z feasible, what broke) and returns a verified written answer with evidence. Read-only end to end, no branch, no code change, no pull request. Use when asked to investigate, diagnose, explain behaviour or assess feasibility, or on /investigate <question>.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# investigate: question in, verified answer out

Read-only, end to end. If the fix becomes obvious mid-flight, the finding
becomes the spec for the task funnel; you do not edit files here. An
investigation that quietly turns into a change skips triage, tests and review,
which is every gate the funnel has.

Input: the question after the command. Empty, ask what to investigate and stop.

## Stage 1 · Sharpen the question

Restate the question and state **what evidence would settle it**. If you cannot
say what would settle it, ask and stop. Fanning out on a vague question buys file
dumps, not an answer.

## Stage 2 · Fan out, read-only, in parallel, in one message

Each subagent gets the full question, because it inherits nothing, plus one
angle:

- **code path**, where it lives, who calls it, what it returns
- **data**, schema, migrations, what the rows actually allow
- **history**, `git log` and `git blame`, the tracker, why it is the way it is
- **runtime**, config, environment, scheduled jobs, webhooks, external services

Skip the angles the question obviously does not touch. Two is a normal number,
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add gabrielbsandim/task-funnel/investigate@sha256:1595d281ecaf99e0

The line at the top of this page pins the label git:20261020.5ec3e90, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_kfcwkffqns3t4azy.svg)](https://markdownregistry.com/a/art_kfcwkffqns3t4azy)

1 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 2 versions of this file, between Aug 17, 2026 and Oct 20, 2026, about one every 64 days.

versioncommittedcommitsizeaudit
git:20261020.5ec3e90 latest2026-10-20 5ec3e90 2,770 BA view · diff
git:20260817.7b6963e2026-08-17 7b6963e 2,709 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (2770 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

gabrielbsandim/task-funnel · 3 stars · license MIT · pushed 2026-10-21 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_kfcwkffqns3t4azy
GET https://markdownregistry.com/api/v1/resolve?ref=gabrielbsandim/task-funnel/investigate
GET https://markdownregistry.com/api/v1/blob/1595d281ecaf99e0610160900b014c795a7b8aa0a8a3c6a98a2be6bef9ced97e

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from gabrielbsandim/task-funnel

backlog SKILL.md skill
gabrielbsandim/task-funnel · skills/backlog/SKILL.md · Grooms an open backlog before the next task is picked: finds issues that are the same work, adopts orphans into the…
git:20261020.5ec3e90 · audit A · 3 stars
note SKILL.md skill
gabrielbsandim/task-funnel · skills/note/SKILL.md · Writes one durable fact into a markdown knowledge vault, with the frontmatter and the provenance that keep it…
git:20261020.377b6fe · audit A · 3 stars
ship SKILL.md skill
gabrielbsandim/task-funnel · skills/ship/SKILL.md · Takes work that is already implemented and committed on a branch and ships it: runs the repo's ship gates off receipts…
git:20261020.5ec3e90 · audit A · 3 stars
task SKILL.md skill
gabrielbsandim/task-funnel · skills/task/SKILL.md · Delivery funnel for one repository task: a feature, bug fix, refactor or chore, from description or issue number to a…
git:20261021.c181593 · audit A · 3 stars

Every file in gabrielbsandim/task-funnel

Other files named investigate

investigate SKILL.md skill
garrytan/gstack · investigate/SKILL.md · Systematic debugging with root cause investigation. (gstack)
v1.0.0 · audit A · 136,966 stars
investigate SKILL.md skill
dcouple/pane · main/src/services/paneChatBundle/skills/investigate/SKILL.md · Investigate a bug or broken behavior through hypothesis-driven root cause analysis and report what is wrong without…
git:20260924.17234b5 · audit A · 577 stars
investigate SKILL.md skill
greenfield-inc/pane · main/src/services/paneChatBundle/skills/investigate/SKILL.md · Investigate a bug or broken behavior through hypothesis-driven root cause analysis and report what is wrong without…
git:20260924.17234b5 · audit A · 575 stars
investigate SKILL.md skill
oliver-kriska/claude-elixir-phoenix · plugins/elixir-phoenix/skills/investigate/SKILL.md · Find the root cause of Elixir/Phoenix bugs — crashes, exceptions, stack traces, compile errors, LiveView that won't…
git:20260929.c08ad14 · audit A · 569 stars
investigate SKILL.md skill
tobihagemann/turbo · claude/skills/investigate/SKILL.md · Systematically investigate bugs, test failures, build errors, performance issues, or unexpected behavior by cycling…
git:20261006.82b4524 · audit A · 414 stars
investigate SKILL.md skill
tobihagemann/turbo · codex/skills/investigate/SKILL.md · Systematically investigate bugs, test failures, build errors, performance issues, or unexpected behavior by cycling…
git:20261006.82b4524 · audit A · 414 stars
investigate SKILL.md skill
testdouble/han · han-coding/skills/investigate/SKILL.md · Evidence-based investigation of issues, bugs, API calls, integrations, and other aspects of software development that…
git:20260915.ec2cb7c · audit A · 319 stars
investigate SKILL.md skill
borda/ai-rig · plugins/codex-rig/skills/investigate/SKILL.md · Investigate code debugging and root-cause narrowing; use measurable gates before fixes.
git:20261009.e7932ee · audit A · 28 stars

Browse by kind, by grade A, or by owner.