subagent-pre-existing-misattribution

subagent-pre-existing-misattribution is a skill for Claude Code from wan-huiyan/agent-traffic-control. It costs 350 tokens per session (2,644 once invoked), scanned A, original, MIT.

A testing guide for telling apart failures caused by a current change from failures that already existed. In multi-agent development, separate agents may pass an assumed baseline of failures from one task to the next.

In plain words
What is it for?
Use it when reviewing subagent work, investigating repeated test failures, or checking whether a reported baseline is still valid.
Why use it?
It prevents real regressions from being dismissed as unrelated and helps catch missed updates to tests that depend on changed lists or counts.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter. Also seen: mentions subagents; mentions Claude Code.

Part of the agent-traffic-control plugin — 107 skills shipped together

Good fit Use it when reviewing subagent work, investigating repeated test failures, or checking whether a reported baseline is still valid.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution
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 wan-huiyan/agent-traffic-control --skill subagent-pre-existing-misattribution
Clone the repo
git clone --depth 1 https://github.com/wan-huiyan/agent-traffic-control

Made for: Claude Code.

Or install agent-traffic-control, the plugin that ships this one along with the rest of its 107 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.

agentmods badge for subagent-pre-existing-misattribution

README.md
[![agentmods](https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution/github.svg)](https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution)
Your own site
<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution/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 subagent-pre-existing-misattribution

Your own site · 80×15
<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/subagent-pre-existing-misattribution.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 350 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,644 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 $0.00350 $0.02644
Opus 5 $0.00175 $0.01322
Sonnet 5 $0.00070 $0.00529
Haiku 4.5 $0.00035 $0.00264

Measured 7d ago against content hash 3b9fa4bfda04, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

subagent-pre-existing-misattribution 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 7d 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.

plugins/agent-traffic-control/skills/subagent-pre-existing-misattribution/SKILL.md · 180 lines

How it starts

The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Subagent-Driven Development: Pre-Existing Failure Misattribution

Problem

In subagent-driven development, each task's reviewer evaluates spec compliance + code quality for THAT task in isolation. When the implementer or reviewer runs the full suite and finds N failures, they look up whether those failures are "new from this task" or "pre-existing baseline." If they classify them wrong, the failures get carried forward as "known unrelated failures" through every subsequent task — and the final architectural reviewer is the only checkpoint that catches the misattribution.

The specific failure mode: an early task removes an item from a list (e.g., a sidebar nav entry, a collection of dashboard cards). Other test files have HARDCODED COUNT ASSERTIONS for that list (assert count == 5) that nobody notices because:

  1. The implementer of the breaking task fixes the parametrize lists they SEE but misses count assertions in OTHER test files.
  2. The implementer reports "8 failures, all pre-existing — verified on clean tree."
  3. Subsequent task implementers/reviewers see "baseline = 8 failures" and treat anything matching that count as expected.
  4. The architectural final reviewer compares against the actual mainline branch (not the branch tip pre-commit) and discovers N of those "pre-existing" failures are actually caused by the early task.

Context / Trigger Conditions

  • Multi-task plan executed via superpowers:subagent-driven-development or similar
  • Reviewer report contains "N pre-existing failures, verified unrelated to this task"
  • Reviewer cites git stash or "clean tree check" as verification mechanism
  • Failure count is suspiciously stable across consecutive tasks (e.g., "8 pre-existing" through Tasks 7, 8, 9, 10 — even after the suite grew)
  • A test file references a count of items in a list that the plan modifies (sidebar tabs, library cards, sub-nav entries, primary routes, etc.)
  • Final architectural review surfaces failures that map directly to changes from earlier tasks

Read the full file on GitHub · 180 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. 7d ago First seen · 180 lines · 350 tokens per session scan A 3b9fa4bfda04

Subscribe to this mod's changes

subagent-pre-existing-misattribution is a skill published in the GitHub repository wan-huiyan/agent-traffic-control (3 stars, last pushed today), licensed MIT. It adds 350 tokens to every session and 2,644 once invoked, about $0.0018 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 2026-09-05.

Related

Other skills, from other repositories

rust-intel

Hard rules for writing Rust in code that already compiles and passes tests but is silently broken, slow, or semver-fragile. Load this BEFORE writing any Rust code. Targets bugs that survive rustc, clippy, and cargo test but fail in production or rot the codebase. Covers async, unsafe, FFI, concurrency, crypto…

PHPCraftdream/rust-intel · 123 tokens

eval-business-logic

Specialized business logic evaluator for the Evaluate-Loop. Use this for evaluating tracks that implement core product logic — pipelines, dependency resolution, state machines, pricing/tier enforcement, packaging. Checks feature correctness against product rules, edge cases, state transitions, data flow, and user…

Ibrahim-3d/orchestrator-supaconductor · 110 tokens

eval-integration

Specialized integration evaluator for the Evaluate-Loop. Use this for evaluating tracks that integrate external services — Supabase auth/DB, Stripe payments, Gemini API, third-party APIs. Checks API contracts, auth flows, data persistence, error recovery, environment config, and end-to-end flow integrity. Dispatched…

Ibrahim-3d/orchestrator-supaconductor · 109 tokens

nazgul:heartbeat

Run one Nazgul automation-heartbeat tick — triages the work inbox and auto-starts the next objective if idle. Opt-in and default-off; fired by an optional Claude Code native scheduled agent (routine) or run by hand. Use when asked to "run a heartbeat tick", "check the inbox", or to test/debug the heartbeat.

OrodruinLabs/nazgul · 74 tokens

qf-fix

Analyze failing tests from the last /qf-run and fix the source code to make them pass. Use when the user runs /qf-fix, asks to "fix failing tests", or asks to "fix test failures". Requires result files from a prior /qf-run.

Qualflare/qualflare-claude-code · 62 tokens

rust-intel

Hard rules for writing Rust in code that already compiles and passes tests but is silently broken, slow, or semver-fragile. Load this BEFORE writing any Rust code. Targets bugs that survive rustc, clippy, and cargo test but fail in production or rot the codebase. Covers async, unsafe, FFI, concurrency, crypto…

PHPCraftdream/rust-intel · 123 tokens