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.
npx agentmods add skills/ayeshakhalid192007-dev/graph-engineering-crash-course/bundle-conflictsnpx skills add ayeshakhalid192007-dev/graph-engineering-crash-course --skill bundle-conflictsgit clone --depth 1 https://github.com/ayeshakhalid192007-dev/graph-engineering-crash-courseWrote 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/skills/ayeshakhalid192007-dev/graph-engineering-crash-course/bundle-conflicts)<a href="https://agentmods.dev/skills/ayeshakhalid192007-dev/graph-engineering-crash-course/bundle-conflicts"><img src="https://agentmods.dev/badge/skills/ayeshakhalid192007-dev/graph-engineering-crash-course/bundle-conflicts.svg" alt="Measured on agentmods" 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 | $0.00043 | $0.00911 |
| Opus 5 | $0.00022 | $0.00456 |
| Sonnet 5 | $0.00009 | $0.00182 |
| Haiku 4.5 | $0.00004 | $0.00091 |
Grade A, and why
bundle-conflicts 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 4d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
bundle-conflicts
Reads a task-scoped subgraph draw plus a list of known unresolved contradictions, and produces a bundle where undisputed context passes through untouched and any contradicting claims stay both present, tagged as an unresolved pair, with the reason they conflict stated in the output.
Instructions
You are a Claude Code skill implementing the conflict-aware-bundle
pattern. Follow these steps in order:
- Read the full task-scoped draw first, noting every node in it.
Default to the draw and known contradiction in this kit's
README.mdunless the user gives you a different one. - Read the list of known unresolved contradictions, noting exactly which node ids each one names.
- Classify every node in the draw as either "touched by a known contradiction" or "undisputed." A node counts as touched only if a known contradiction explicitly names it — do not invent a contradiction for two claims that merely differ in detail but weren't flagged, and do not treat a node as disputed just because it sits near a disputed one in the graph.
- Pass undisputed nodes through unchanged. No conflict tag, no alteration to their value or source — the bundle should not make settled information look uncertain.
- For every contradiction, include every claim it names, never a subset. Do not drop one side because it's labeled "backup," lower- confidence, or secondary in some other field — the pattern exists specifically so a role label like "primary" can't quietly settle a disagreement that hasn't actually been resolved.
- Never compute or report a resolved value for a contradiction — no average, no pick-the-primary-source default, no "most likely" answer. The bundle's job is to surface the disagreement, not settle it.
- Tag each contradiction explicitly, naming the claims involved and stating in plain language why they conflict (what they both claim, and why those claims can't both be true at once).
- Report the bundle in two clearly separated groups: undisputed nodes as-is, then each unresolved-contradiction pair with its tag and reason. A reader should be able to tell at a glance which nodes in the bundle are settled and which are still actively disputed.
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.
- 4d ago First seen · 93 lines · 43 tokens per session scan A ddcc6fa32b7b
bundle-conflicts is a skill published in the GitHub repository ayeshakhalid192007-dev/graph-engineering-crash-course (5 stars, last pushed 9d ago), licensed MIT. It adds 43 tokens to every session and 911 once invoked, about $0.0002 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 skills, from other repositories
rulesync
Generates and syncs AI rule configuration files (.cursorrules, CLAUDE.md, copilot-instructions.md) across 20+ coding tools from a single source. Use when syncing AI rules, running rulesync commands, importing or generating rule files, or managing shared AI coding configurations.
establishing-project-context
Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.
autoprompt
Explicit-only useful-first orchestration. Invoke /autoprompt to turn a mission into one executable roadmap, build dependency-safe lanes, and verify the result with independent reviewers. Never infer invocation from ordinary requests. Never resume from leftover artifacts without an explicit resume instruction.
loop
Full execution protocol for MODE: LOOP — the compound-engineering loop: brainstorm → plan → build → review → improve, iterating under defense-in-depth stop conditions with generator/critic separation, durable resumable state, and mandatory compounding learning capture. Loaded on demand by the architect when the loop…
restore-internals-seams-in-finally-blocks-after-each-test
When delegating a task affected by this skill, include.
loongsuite-pilot-insight
基于 LoongSuite Pilot / AI Coding Agent 日志生成事件洞察、组织洞察、数据质量、研发效能和 AI Native 使用类 SLS 报表时使用；包含 AI Coding 事件表语义，以及团队报表可选的部门维表、deptuser 组织关系、指标口径和公共 CTE，通常与 sls-dashboard-builder 一起使用。.