Atlas / Skills / jeremylongshore / Linear Local Dev Loop

Linear Local Dev LoopBLOCK

skills/jeremylongshore/linear-local-dev-loop

Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.

Verdict
BLOCK
Grade
D
Trust score
69 /100
Version
1.13.0
Hosts
1 documented
License
MIT
Stars
2,824
01

Overview

Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.

Read from source at commit 4f83675ca38aOBSERVED · 2026-10-09
02

Host compatibility

What the documentation claims. We have not run a compatibility test.

HostStatusNotes
claude-codementioned
03

What it tells the agent

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: linear-local-dev-loop
description: >-
  Build a fast Linear integration development loop using fixtures, contract boundaries, and an isolated test workspace. Use when iterating locally without polluting production Linear data. Trigger with "develop Linear locally", "mock Linear GraphQL", or "test Linear webhooks locally".
argument-hint: "[repository-path] [graphql|sdk|webhook]"
allowed-tools: Read, Glob, Grep, WebFetch, Write, Edit
version: 1.13.0
author: Jeremy Longshore <[email protected]>
license: MIT
tags:
- saas
- linear
- local-development
model: inherit
effort: high
compatibility: Designed for Claude Code; live verification requires network access and an approved Linear workspace credential
---
# Linear Local Development Loop

## Overview

Keep routine tests deterministic and offline, then use a deliberate read-only or sandbox-workspace probe to detect contract drift.

## Prerequisites

- The target repository, Linear workspace, environment, and accountable owner
- Current security, privacy, compliance, capacity, and change-control requirements
- An approved Linear credential only when a bounded live verification is necessary

## Tool Discipline

Use `Read`, `Glob`, and `Grep` to inspect code, configuration, and evidence. Use `WebFetch` only for current first-party Linear documentation and package metadata. Use `Write` or `Edit` only for requested implementation with known target files. Never write credentials, customer content, unrestricted environment output, or unredacted GraphQL variables.

## Current Contract

- Webhook delivery requires a publicly accessible HTTPS, non-localhost URL; localhost-only receivers are not a production-equivalent test.
- Signature verification requires the exact raw request body, so body-parser ordering belongs in local contract tests.
- GraphQL introspection and official generated SDK types are better schema authorities than hand-written guessed fixtures.

## Authentication

Use a personal API key only for owner-controlled scripts, OAuth with PKCE for user-delegated applications, or an enabled client-credentials grant for approved automation. Personal keys use `Authorization: <API_KEY>`; OAuth tokens use `Authorization: Bearer <ACCESS_TOKEN>`. Store credentials server-side in an approved secret manager.

Treat app approval, team access, scope changes, credential creation, rotation, revocation, and production access as owner-approved actions.

## Instructions

1. Inspect the repository's test framework, GraphQL client boundary, fixtures, environment loading, and webhook middleware order.
2. Record minimal redacted success, partial-error, auth-error, throttling, pagination, and webhook fixtures.
3. Inject a fake client at the adapter boundary; do not mock every SDK model method across the application.
4. Test raw-body signature verification, duplicate delivery IDs, five-second acknowledgement behavior, and queued processing.
5. Use an isolated workspace and approved tunnel only for explicit end-to-end checks; label and clean up test records.
6. Add a protected contract probe that detects schema/package drift without mutating production.

## Approval Boundaries

Do not create, reveal, rotate, or revoke credentials; authorize an OAuth app; change scopes or team access; create, mutate, archive, or delete workspace data; configure or re-enable webhooks; import or export data; change roles, SCIM, or audit streaming; transmit diagnostics; change paid entitlements; or perform another production mutation without explicit approval from the accountable owner. Keep diagnosis read-only unless implementation was requested.

## Output

Return the workspace and team scope, auth mode without credential value, files and contracts inspected, exact operation names, evidence collected, validation result, sensitive fields redacted, remaining risk, accountable owner, approval state, and rollback or next action.

## Error Handling

| Condition | Response |
|---|---|
| Fixture contains customer data | Replace it with a synthetic equivalent and purge retained copies. |
| Signature fails locally | Confirm the raw bytes reach verification before JSON parsing. |
| Tunnel unavailable | Run signed synthetic requests locally and defer the real-delivery proof. |
| Test reaches production | Fail closed unless the protected live-test gate is explicitly enabled. |

## Examples

Use a compact handoff that makes scope, mutation authority, and verification evidence reviewable.

Input:

```text
adapter=graphql; fixtures=synthetic; webhook=raw-body; live=false
```

Expected handoff:

```text
unit=offline; contract=deterministic; production-writes=blocked
```

## Resources

- [Skill-specific official documentation](references/official-docs.md)
- [Linear developer documentation index](https://linear.app/llms.txt)
- [Linear GraphQL API](https://linear.app/developers/graphql.md)
04

Trust audit

BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)FAIL
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (2)

HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
SKILL.md:16
compatibility: Designed for Claude Code; live verification requires network access and an approved Linear workspace credential
Why it matters. asks the agent to read credentials
HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
SKILL.md:44
Treat app approval, team access, scope changes, credential creation, rotation, revocation, and production access as owner-approved actions.
Why it matters. asks the agent to read credentials

Gates applied: no_behavioural_pass.

Audited 2026-10-09 · audit v0.4.1 · source sha 4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__linear-local-dev-loop.json · Report an issue / request a re-scan
05

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-10-094f83675ca38aBLOCKD69first audit
06

Questions

What does the Linear Local Dev Loop skill do?

Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.

Is Linear Local Dev Loop safe to install?

No — not without reading the findings first. The audit graded it D (69/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on.

What can Linear Local Dev Loop access on my machine?

The audit observed no filesystem, network or shell use at all in its source.

Which assistants does Linear Local Dev Loop work with?

Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.

How current is this page?

The grade is for one exact copy of the source (4f83675ca38a), read on 2026-10-09. The repository is watched, and a new audit runs when it changes — this is the first audit.

Advertisement