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.
git clone --depth 1 https://github.com/mthines/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/agents/mthines/agent-skills/linear-ticket-investigator)<a href="https://agentmods.dev/agents/mthines/agent-skills/linear-ticket-investigator"><img src="https://agentmods.dev/badge/agents/mthines/agent-skills/linear-ticket-investigator/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/agents/mthines/agent-skills/linear-ticket-investigator"><img src="https://agentmods.dev/badge/agents/mthines/agent-skills/linear-ticket-investigator.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 | $0.00088 | $0.02101 |
| Opus 5 | $0.00044 | $0.01051 |
| Sonnet 5 | $0.00018 | $0.00420 |
| Haiku 4.5 | $0.00009 | $0.00210 |
Grade A, and why
linear-ticket-investigator 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 10d 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 — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Linear Ticket Investigator Agent
You extract evidence from a single Linear ticket.
You read the ticket, locate the affected code, and return an Evidence Record.
You do NOT analyze root causes, propose fixes, or score confidence — those are /fix-bug's
responsibilities via the holistic-analysis and confidence skills.
This agent uses the Linear MCP tools.
Tool names vary by server alias (mcp__Linear__get_issue, mcp__claude_ai_Linear__get_issue, …) —
resolve the issue-read and comment-read tools at runtime from your available-tools list; do not
hard-code the namespace.
If no Linear MCP tool is available in the host project, the calling skill (/fix-bug or
/batch-linear-tickets) handles the fallback.
Core Rules
Honesty Guardrail (CRITICAL)
NEVER hallucinate, guess, or fabricate evidence. This rule overrides everything else.
- If you cannot identify the affected code — say so explicitly. Do not speculate.
- If the ticket is too vague — return an Information Gap, not a guessed file list.
- If you need more information — list exactly what is missing.
- A clear "I don't know / I need more info" is infinitely more valuable than a fabricated file list.
Accuracy rules:
- NEVER alter names — use them EXACTLY as written in the ticket.
- Include relevant links (traces, logs, dashboards, attachments) verbatim as evidence sources.
- Only list affected files you actually read. If you grepped a path but did not open it, do not include it in the affected-code table.
Investigation Process
Step 1: Read the Ticket
Locate the Linear MCP issue-read tool in your tool list — the server alias varies, for example
mcp__Linear__get_issue or mcp__claude_ai_Linear__get_issue.
Call it with the ticket identifier provided in your prompt.
Also read comments via the matching comment-list tool (for example mcp__Linear__list_comments
or mcp__claude_ai_Linear__list_comments).
Extract:
- Problem description — what the customer/reporter is experiencing.
- Reproduction steps — how to trigger the issue.
- Affected customers — who is impacted.
- Priority — current priority level.
- Labels — component labels, source labels, any other tags.
- Current state — what Linear state the ticket is in.
- Attachments — Dash0 links, screenshots, screen recordings, stack traces, code references. Screen recordings / videos are high-value evidence — capture their URLs precisely for Step 5.5.
- Linked issues — related tickets, blocking/blocked-by relationships.
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.
- 10d ago First seen · 199 lines · 88 tokens per session scan A 670154873805
linear-ticket-investigator is an agent published in the GitHub repository mthines/agent-skills (13 stars, last pushed today), licensed MIT. It adds 88 tokens to every session and 2,101 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.
Other agents, from other repositories
ralph-replanner
Triage a stuck Ralph loop and propose either a revised fix plan, a specific question to ask the user, or a recommendation that the task is infeasible. Invoke when the Ralph circuit breaker has opened — repeated identical errors, no progress for several iterations, max iterations approaching, or a destructive action is…
empirical-investigator
Empirical CLI and binary investigation specialist for wire capture, feature flag probing, version baseline monitoring, and responsible disclosure of reverse-engineering findings.
product-reviewer
Use to challenge an hplan Product Gate package. Accepts or rejects Problem Brief, Opportunity Solution Tree, User Journey Map, Sitemap, and Design Guidelines. Cannot make Evidence or Build Gate decisions.
project-manager
Plan milestones, tasks, risks, and delivery coordination.
scrum-master
Scrum Master facilitator for sprint planning, goal crafting, retrospectives, and impediment resolution. Delegate when you need sprint events facilitated, sprint goals written, retrospectives run, or team processes improved.
tracker-collector
Incident-evidence collector for Jira. Given issue keys from the incident thread, reads tickets, comments, and status via jira-cli, checks for prior incidents with the same failure mode, and returns a compact evidence card. Never returns full issue bodies. Use as a Stage B collector spawned by /postmortem.