Skills loki-mode

📦

loki-mode

Name: loki-mode
Author: sickn33

Content revision r1 Critical ⚙️ External commands🌐 Network access📁 Filesystem access⚡ Contains scripts🔑 Env variables

Orchestrate Multi-Agent Product Delivery

Complex product work can require coordination across planning, coding, testing, and operations. Loki Mode provides a structured multi-agent workflow with task queues, checkpoints, and review steps.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Do not auto-install this skill.

The canonical policy requires operator review before any installation action.

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "loki-mode" from https://skillstore.io/skills/sickn33-loki-mode.md and its manifest at https://skillstore.io/api/skills/sickn33-loki-mode/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/sickn33-loki-mode.md Signed manifest GET /api/skills/sickn33-loki-mode/manifest Signed lockfile GET /api/skills/sickn33-loki-mode/lockfile

Test it

Using "loki-mode". Create a plan for an authenticated reporting endpoint.

Expected outcome:

Discovery tasks for data and access requirements.
Implementation tasks for the endpoint and tests.
A review checkpoint before database or deployment changes.

Using "loki-mode". Prepare a release-readiness workflow.

Expected outcome:

A prioritized checklist for tests, documentation, and monitoring.
Assigned review stages with clear completion criteria.
An approval gate before publishing or deployment.

Security Audit

Critical

v5 • 7/21/2026 Open versioned report

Static destructive-command, screen-capture, weak-cryptography, environment, and long-line alerts are false positives: they are blocklist text, documentation, local demo tooling, or benchmark artifacts. However, the skill explicitly requires a permission bypass and directs unsupervised execution without confirmations; it also contains a dashboard HTML injection risk. Publishing should remain blocked until these issues are remediated. Static review was capped at 400/5857 representative findings; omitted static matches are unconfirmed, so automatic publishing stays disabled until manual review.

1082

Files scanned

72,871

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (2)

Critical

SKILL.md:93-110

Unsupervised execution and permission-bypass directive

The skill instructs users to launch with a permission-bypass flag and directs agents never to ask questions, wait for confirmation, or stop voluntarily. This removes essential approval controls before file, command, deployment, and business actions.

The cited installation and core rules explicitly require permission bypass and zero human intervention. The unsafe operating intent is direct and unambiguous.

High

autonomy/run.sh:714-745

Unescaped dashboard data inserted into HTML

The dashboard interpolates agent and task values into HTML strings assigned through innerHTML without escaping. Malicious values in local state JSON can execute script in the dashboard context.

The cited render functions concatenate agent and task fields into HTML, and later dashboard code assigns generated strings with innerHTML. No escaping is visible in the cited path.

Risk Factors

⚙️ External commands (50)

autonomy/CONSTITUTION.md:40 autonomy/run.sh:725-739 autonomy/run.sh:744 autonomy/run.sh:745 autonomy/run.sh:56 autonomy/run.sh:57 autonomy/run.sh:163 autonomy/run.sh:174 autonomy/run.sh:184 autonomy/run.sh:194 autonomy/run.sh:202 autonomy/run.sh:294 autonomy/run.sh:296 autonomy/run.sh:321 autonomy/run.sh:326 autonomy/run.sh:327 autonomy/run.sh:328 autonomy/run.sh:329 autonomy/run.sh:336 autonomy/run.sh:810 autonomy/run.sh:840 autonomy/run.sh:841 autonomy/run.sh:844 autonomy/run.sh:853 autonomy/run.sh:854 autonomy/run.sh:855 autonomy/run.sh:856 autonomy/run.sh:857 autonomy/run.sh:858 autonomy/run.sh:859 autonomy/run.sh:861 autonomy/run.sh:862 autonomy/run.sh:863 autonomy/run.sh:866 autonomy/run.sh:896 autonomy/run.sh:952 autonomy/run.sh:960-961 autonomy/run.sh:961 autonomy/run.sh:1018 autonomy/run.sh:1022 autonomy/run.sh:1026 autonomy/run.sh:1037 autonomy/run.sh:1046-1047 autonomy/run.sh:1222 autonomy/run.sh:1225 autonomy/run.sh:1226 autonomy/run.sh:1246 autonomy/run.sh:1254 autonomy/run.sh:1255 autonomy/run.sh:1259

🌐 Network access (50)

📁 Filesystem access (50)

autonomy/README.md:187 autonomy/README.md:187 autonomy/run.sh:749 autonomy/run.sh:750 autonomy/run.sh:751 autonomy/run.sh:752 autonomy/run.sh:753 autonomy/run.sh:777 autonomy/run.sh:1072 autonomy/run.sh:1136 autonomy/run.sh:243 autonomy/run.sh:1013 autonomy/run.sh:1039 autonomy/run.sh:1057 autonomy/run.sh:1072 autonomy/run.sh:1136 autonomy/run.sh:79 autonomy/run.sh:162 autonomy/run.sh:163 autonomy/run.sh:173 autonomy/run.sh:183 autonomy/run.sh:193 autonomy/run.sh:201 autonomy/run.sh:208 autonomy/run.sh:217 autonomy/run.sh:294 autonomy/run.sh:321 autonomy/run.sh:326 autonomy/run.sh:327 autonomy/run.sh:328 autonomy/run.sh:329 autonomy/run.sh:373 autonomy/run.sh:374 autonomy/run.sh:810 autonomy/run.sh:939 autonomy/run.sh:940 autonomy/run.sh:1175 autonomy/run.sh:1177 autonomy/run.sh:1193 autonomy/run.sh:1199 autonomy/run.sh:1211 autonomy/run.sh:1212 autonomy/run.sh:1246 autonomy/run.sh:1306 autonomy/run.sh:1307 autonomy/run.sh:1328 autonomy/run.sh:1333 autonomy/run.sh:1364 autonomy/run.sh:1377 autonomy/run.sh:1390

⚡ Contains scripts (50)

🔑 Env variables (45)

Audited by: claude View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/sickn33-loki-mode/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/sickn33-loki-mode/security.svg)](https://skillstore.io/skills/sickn33-loki-mode?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/sickn33-loki-mode?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/sickn33-loki-mode/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/sickn33-loki-mode.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

sickn33. (2026). loki-mode security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/sickn33-loki-mode/audits/5

BibTeX citation

@techreport{sickn33-sickn33-loki-mode-2026,
  author = {sickn33},
  title = {loki-mode security audit report (audit version 5)},
  institution = {Skillstore},
  year = {2026},
  number = {5},
  url = {https://skillstore.io/skills/sickn33-loki-mode/audits/5},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "loki-mode security audit report (audit version 5)"
version: "unspecified"
type: report
authors:
  - name: "sickn33"
date-released: "2026-07-21"
url: "https://skillstore.io/skills/sickn33-loki-mode/audits/5"
identifiers:
  - type: other
    value: "skillstore:sickn33-loki-mode:audit:5"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Plan a new service

Turn a product brief into phased architecture, development, testing, and release tasks.

Coordinate release readiness

Organize review, test, documentation, and deployment preparation tasks in one workflow.

Evaluate agent workflows

Use the local dashboard and benchmark materials to study multi-agent coordination patterns.

Try These Prompts

Start a small workflow

Loki Mode, create a plan for a small API feature. Stop after the plan and ask for approval before making changes.

Prepare release tasks

Loki Mode, organize release-readiness tasks for this repository. Include tests, documentation, and a human approval checkpoint.

Coordinate a feature

Loki Mode, break this approved feature into scoped implementation and review tasks. Keep all changes inside the repository and require approval before deployment.

Review workflow controls

Loki Mode, audit this workflow plan for unsafe permissions, external actions, and missing approval gates. Produce recommendations only.

Best Practices

Keep workflows scoped to a repository and approved objective.
Require human approval before deployments, external messages, or permission changes.
Set conservative agent limits and verify each completed task.