gh-issue-recon

gh-issue-recon is a command for Claude Code from skyfox675/agents-skills. It costs 117 tokens per session (1,259 once invoked), scanned A, original, MIT.

A read-only GitHub issue investigation command for checking whether a planned issue is understood and technically workable. GitHub issues are entries used to describe work, bugs, or feature requests.

In plain words
What is it for?
Use it to inspect one or more groomed issues, trace each request through the codebase, find relevant file and line references, and report how the work could be implemented.
Why use it?
It replaces guesswork about implementation size and affected code with a checked approach, effort estimate, risks, and dependencies.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions subagents.

Good fit Use it to inspect one or more groomed issues, trace each request through the codebase, find relevant file and line references, and report how the work could be implemented.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/skyfox675/agents-skills/gh-issue-recon
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.

Clone the repo
git clone --depth 1 https://github.com/skyfox675/agents-skills

Made for: Claude Code.

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.

agentmods badge for gh-issue-recon

README.md
[![agentmods](https://agentmods.dev/badge/commands/skyfox675/agents-skills/gh-issue-recon/github.svg)](https://agentmods.dev/commands/skyfox675/agents-skills/gh-issue-recon)
Your own site
<a href="https://agentmods.dev/commands/skyfox675/agents-skills/gh-issue-recon"><img src="https://agentmods.dev/badge/commands/skyfox675/agents-skills/gh-issue-recon/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.

agentmods 80×15 button for gh-issue-recon

Your own site · 80×15
<a href="https://agentmods.dev/commands/skyfox675/agents-skills/gh-issue-recon"><img src="https://agentmods.dev/badge/commands/skyfox675/agents-skills/gh-issue-recon.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 117 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,259 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00117 $0.01259
Opus 5 $0.00059 $0.00629
Sonnet 5 $0.00023 $0.00252
Haiku 4.5 $0.00012 $0.00126

Measured 10d ago against content hash fd79529e9798, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

gh-issue-recon 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

.claude/commands/gh-issue-recon.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/gh-issue-recon — technically scope, size, and de-risk groomed issues

Operate in caveman mode (load the caveman skill) — recon fans out and tokens compound. Keep gh commands, labels, file:line refs, and the TECH-RECON contract line byte-exact; the findings comment is the exception (humanizer — see below).

Arguments: $ARGUMENTS

Parse them as:

  • One or more issue numbers (with or without #). Each is a separately groomed, stakeholder-accepted issue that the dev team has not yet technically reviewed or sized. A lead may drop several at once.
  • Optional model:<tier> and effort:<level> tokens, anywhere — these pin the recon agent's tier (default sonnet; escalate to opus for deep or cross-cutting asks). They are NOT the implementation tier — recon recommends that separately in its findings. The cheap tier is never used here (technical judgement on real code needs a capable model; see the technical-recon skill).

Example: /gh-issue-recon 412 415 418 model:opus

The technical-recon skill is the protocol — the read-only sandbox, the findings-comment anatomy (implementation approach, verified file:line, LoE + confidence, risks/unknowns, dependencies, split recommendation, dispatch recommendation), the verdict states, and the output contract all come from it. This is the deeper dev-side pass, distinct from the cheap intake-triage recon in the gh-issue-filing skill. Consult it before dispatching.

Steps

  1. Resolve the batch. Collect the issue numbers from the args. Strip the model:/effort: tokens first. For each, confirm it exists and is open (gh issue view <N> --json title,body,labels,state). Skip and report any that are closed or carry an operator hold (do-not-dispatch, per gh-issue-labels).

  2. Fan out, one read-only recon agent per issue, in parallel (single message, per the dispatching-subagents skill). Each agent runs at the resolved tier, briefed with the technical-recon sandbox verbatim: it first reads any prior recon questions + answers already on the issue (a rerun resumes, it does not restart), then reads the repo and the issue — but writes no code, opens no PR, and never claims/assigns the issue (claiming is dispatch-time). Pass each agent the conflict-zone list (files other operators' open PRs touch) so its approach avoids them.

Read the full file on GitHub · 37 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.

  1. 10d ago First seen · 37 lines · 117 tokens per session scan A fd79529e9798

Subscribe to this mod's changes

gh-issue-recon is a command published in the GitHub repository skyfox675/agents-skills (10 stars, last pushed 9d ago), licensed MIT. It adds 117 tokens to every session and 1,259 once invoked, about $0.0006 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.