A failure-to-code lookup tool that starts with a stack trace, failed test, or code change and identifies the relevant source locations. It ranks the likely areas to inspect.

Use it to investigate stack traces, failing test IDs, error messages, or changed files and find the code they implicate.

It gives you a focused starting point when an error or test failure does not clearly identify the real cause.

glia-debug — Skill for Claude Code by James-Chahwan/repo-graph

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.

Any agent

npx skills add James-Chahwan/repo-graph --skill glia-debug

Clone the repo

git clone --depth 1 https://github.com/James-Chahwan/repo-graph

Made for: Claude Code.

Or install repo-graph, the plugin that ships this one along with the rest of its 8 skills.

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/james-chahwan/repo-graph/glia-debug/github.svg)](https://agentmods.dev/skills/james-chahwan/repo-graph/glia-debug)

Your own site

<a href="https://agentmods.dev/skills/james-chahwan/repo-graph/glia-debug"><img src="https://agentmods.dev/badge/skills/james-chahwan/repo-graph/glia-debug/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/james-chahwan/repo-graph/glia-debug"><img src="https://agentmods.dev/badge/skills/james-chahwan/repo-graph/glia-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 894 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.

Model	Per session	Once invoked
Fable 5.1	$0.00054	$0.00894
Opus 5.5	$0.00022	$0.00358
Sonnet 5.5	$0.00011	$0.00179
Haiku 4.5	$0.00005	$0.00089

Measured 5d ago against content hash 302e7dbd3f1c, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

glia-debug 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 5d 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.

claude-plugin/skills/glia-debug/SKILL.md · 39 lines

How it starts

The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debug from a failure

Start from the error, not a filename. glia matches the frames, test id or changed paths in a failure to nodes in its graph, then ranks the code around them.

Transports

Use whichever is available, in this order: the repo-graph MCP tools when loaded (mcp__repo-graph__<tool>, or mcp__plugin_repo-graph_repo-graph__<tool> from the Claude Code plugin; they may be deferred, so search your tool list first); the glia CLI when on PATH; else uvx --from glia-cli glia <command> .. (no install; the first run downloads the wheel); else uvx mcp-repo-graph <verb>. The MCP tools and uvx mcp-repo-graph print the same text; the glia CLI answers the same question with --json rows (file, line, tier). Run any of them at the repo root (. below).

Want	MCP tool call	glia CLI	uvx mcp-repo-graph
Code a stacktrace points at	`find query=<failure> kind=stacktrace`	`glia resolve . "<signal>" --kind stacktrace --json`	`uvx mcp-repo-graph find - --kind stacktrace`
Code a failing test id points at	`find query=<test id> kind=test`	`glia resolve . "<test id>" --kind test --json`	`uvx mcp-repo-graph find "<test id>" --kind test`
What a diff touches	`find query=<diff> kind=diff`	`glia diff-impact . --diff <file> --json`	`uvx mcp-repo-graph find - --kind diff`
The tests to run	`impact nodes=<node>`	`glia tests-for . <node> --json`	`uvx mcp-repo-graph impact "<node>"`

Steps

Give the raw failure. Don't trim it to one line: the frames are the signal. --kind is auto (sniff, the default), stacktrace, test or diff.

glia resolve . "$(pytest -x 2>&1 | tail -n 80)" --kind stacktrace --json
glia resolve . "tests/test_cart.py::test_total" --kind test --json
git diff > /tmp/d.diff && glia diff-impact . --diff /tmp/d.diff --json
pytest -x 2>&1 | uvx mcp-repo-graph find - --kind stacktrace

Read the top hits: open each row's file:line (over MCP, read node=<a>,<b> slices them in one call with a context: footer of callers and covering tests).
Find the tests that cover the suspect: glia tests-for . <node> --json, or for the whole change glia tests-for . --base main --json. Run those first.
Widen only if the cause isn't there: glia blast-radius . <node> --direction both --depth 1 --json lists the direct callers and callees.

Read the full file on GitHub · 39 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.

5d ago First seen · 39 lines · 54 tokens per session scan A 302e7dbd3f1c

Subscribe to this mod's changes

glia-debug is a skill published in the GitHub repository James-Chahwan/repo-graph (55 stars, last pushed 3d ago), licensed MIT. It adds 54 tokens to every session and 894 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-02.

Debugging

glia-debug next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
glia-debugJames-Chahwan/repo-graph	Skill	55	3d ago	MIT	A	54	Claude Code
systematic-debuggingopen-metadata/OpenMetadata	Skill	16k	yesterday	Apache-2.0	A	37	Claude Code
log-error-digestzebbern/claude-code-guide	Skill	4.7k	3d ago	MIT	A	71	Claude Code, Codex
cwv-optimizeradobe/skills	Skill	319	yesterday	Apache-2.0	A	112	Claude Code
ci-fixersipyourdrink-ltd/bernstein	Skill	1.2k	yesterday	Apache-2.0	A	15	Claude Code, Codex
incident-response-incident-responsermyndharis/antigravity-skills	Skill	1.9k	23d ago	MIT	A	36	Claude Code
triaging-web-analytics-supportPostHog/posthog	Skill	41k	yesterday	none	A	151	Claude Code, Codex