Agent skill · Audited from source
coderunner
skill · instavm/coderunnerApache-2.0 licenceAudited 12 Nov 2026 · source 132da8183323
Is it safe? With care — CAUTION, grade B (87/100). We read this skill's source at commit 132da8183323 and observed filesystem and network access and a hard-coded secret (see the findings). The audit raised 13 findings, 0 critical or high. What it does: Image processing skill for cropping images to 50% from center and rotating them 90 degrees clockwise. This skill should be used when users request image cropping to center, image rotation, or both operations combined on…
Cite
Cite this page
ModelsAtlas. “coderunner (instavm/coderunner)”. Retrieved 12 Nov 2026. https://modelsatlas.com/skills/instavm/coderunner
Trust audit
Audit 0.4.1 · source read 12 Nov 2026
CAUTION means the audit found things worth knowing before you trust this skill. Grade B is shared by 17,406 of 18,389 audited skills.
What the source does
- FilesystemDeclared3 observations
- NetworkDeclared12 observations
- ShellNone observedNo commands run in the source
- DependenciesNot pinnedSome versions float
- SecretsFoundSee the findings; values are never shown
Five audit layers
| Layer | What we check | Result | Findings |
|---|---|---|---|
L0 Provenance | Repo owner, licence, archive status, activity | PASS | 1 |
L1 Static source | Secrets, path traversal, supply chain, hidden code | WARN | 9 |
L2 Instructions | Prompt injection, authority framing, hidden directives | PASS | 3 |
L3 Inventory | The files and declarations match the source | PASS | 0 |
L4 Sandbox | Run in isolation and watch behaviour | SKIPPED why | 0 |
Gates applied: no behavioural pass.
Findings (13)
MEDIUMNetwork egressnet.raw_ip · server.py:60
"http://127.0.0.1:*",
- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMNetwork egressnet.raw_ip · server.py:76
JUPYTER_HTTP_URL = os.environ.get("CODERUNNER_JUPYTER_URL", "http://127.0.0.1:8888").rstrip("/")- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMHard-coded secretssecret.generic · tests/test_auth.py:12
self.token = "…"
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWFilesystem / pathfs.traversal · test-e2e.sh:87
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"get_skill_file","arguments":{"skill_name":"..","filename":"../../../../etc/passwd"}}}')- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.metadata_ip · tests/test_server.py:58
"http://169.254.169.254/latest/meta-data/", "http://192.168.64.1:8223/", "http://[::1]/",
- Why
- cloud metadata endpoint: the classic SSRF credential grab
- Fix
- —
LOWNetwork egressnet.raw_ip · DOCKER_SETUP.md:48
curl -H "Authorization: Bearer $CODERUNNER_API_TOKEN" http://127.0.0.1:8222/mcp
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.raw_ip · DOCKER_SETUP.md:134
curl -f -H "Authorization: Bearer $CODERUNNER_API_TOKEN" http://127.0.0.1:8222/mcp || echo "MCP not ready"
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.raw_ip · README.md:83
In this mode, the MCP server is available at `http://127.0.0.1:8222/mcp`. The setting is fixed when the container is created; the installer refuses to resume a container with a different network mode.
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWSupply chainsupply.unpinned · examples/requirements.txt
openai-agents
- Why
- 1 requirement(s) not pinned with ==
- Fix
- pin exact versions
LOWPrompt injectionprompt.authority_framing · README.md:243
3. Start executing Python code with full access to the sandboxed environment
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWPrompt injectionprompt.authority_framing · THIRD_PARTY_NOTICES.md:101
All rights granted under this License are granted for the term of copyright on the Program, and are irrevocable provided the stated conditions are met. This License explicitly affirms your unlimited p
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWPrompt injectionprompt.authority_framing · THREAT_MODEL.md:61
- Several users with different permissions. The token is one credential with full access.
- Why
- No further detail recorded for this rule.
- Fix
- —
INFOInventory / provenanceinv.oversize · images/runcode.png
images/runcode.png
- Why
- 8392982 bytes not read
- Fix
- —
Audit 0.4.1 · source sha 132da8183323 · 12 Nov 2026Report an issue or request a re-scan
Provenance
Who published it, and what the verdict is pinned to
- Publisherinstavm GitHub organizationOwner of the repository this skill is read from
- Source folderinstavm/coderunnerThe exact folder we audited
- LicenceApache-2.0As the repository states it
- Commit pinned for this verdict132da8183323A new audit runs when the source changes
- Repository statusNot archivedLast push 2 days ago
- Sandboxed runNot runSkills are read, not executed; layer L4 is skipped
What it tells the agent
Full source on GitHub ↗Its description: Image processing skill for cropping images to 50% from center and rotating them 90 degrees clockwise. This skill should be used when users request image cropping to center, image rotation, or both operations combined on…
---
name: image-crop-rotate
description: Image processing skill for cropping images to 50% from center and rotating them 90 degrees clockwise. This skill should be used when users request image cropping to center, image rotation, or both operations combined on image files.
---
# Image Crop and Rotate
## Overview
This skill provides functionality to crop images to 50% of their original size from the center and rotate them 90 degrees clockwise. It uses a reliable Python script with PIL/Pillow for consistent, high-quality image processing.
## When to Use This Skill
Use this skill when the user requests:
- Cropping an image to 50% from the center
- Rotating an image 90 degrees (clockwise)
- Both cropping and rotating an image
- Image processing tasks that combine center cropping with rotation
## How to Use This Skill
### Overview
The skill provides a single script that performs both operations: cropping to 50% from center and rotating 90 degrees clockwise.
### Process
1. **Identify the input image**: Locate the user's uploaded image file in `/mnt/user-data/uploads/`
2. **Execute the script**: Run the `crop_and_rotate.py` script with input and output paths:
```bash
python scripts/crop_and_rotate.py <input_path> <output_path>
```
3. **Provide the result**: Move the processed image to `/mnt/user-data/outputs/` and share it with the user
### Script Details
**`scripts/crop_and_rotate.py`**
This script performs two operations in sequence:
1. Crops the image to 50% of its original size, centered
2. Rotates the cropped image 90 degrees clockwise
**Usage:**
```bash
python scripts/crop_and_rotate.py input.jpg output.jpg
```
**Arguments:**
- First argument: Path to input image
- Second argument: Path to save processed image
**Supported formats:** Any format supported by PIL/Pillow (JPEG, PNG, GIF, BMP, TIFF, etc.)
**Output:** The script prints processing details including original size, cropped size, and final size
### Example Workflow
```bash
# Process an uploaded image
python /mnt/user-data/skills/image-crop-rotate/scripts/crop_and_rotate.py \
/mnt/user-data/uploads/photo.jpg \
/mnt/user-data/outputs/photo_processed.jpg
```
The script will:
1. Open the input image
2. Crop it to 50% from the center (e.g., 1000x800 → 500x400)
3. Rotate the cropped image 90° clockwise (e.g., 500x400 → 400x500)
4. Save the result to the output pathExcerpt, first 2,395 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.
Install
Pinned to the audited commit 132da81
git clone https://github.com/instavm/coderunner.git pip install -r examples/requirements.txt claude mcp add --transport http coderunner http://coderunner.local:8222/mcp \
git clone https://github.com/instavm/coderunner git -C coderunner checkout 132da8183323
Commands as the repository documents them: shown, never run by us. Our verdict covers commit 132da8183323 only.
Works in which agents?
What its own text says, not a test
| Agent | Status | More |
|---|---|---|
| Claude Code | Mentioned in its text, not tested | Skills for Claude Code · About Claude Code |
| Claude Desktop | Mentioned in its text, not tested | Skills for Claude Desktop · About Claude Desktop |
| Gemini CLI | Mentioned in its text, not tested | Skills for Gemini CLI · About Gemini CLI |
Audit history
Every audit, every source change
- !CAUTION B 87 / 100
first audit · source
132da8183323
README badge
For maintainers · shows the latest audit
[](https://modelsatlas.com/skills/instavm/coderunner/)
The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.
Similar skills
All devops & cloud skills →Th0rgal/sandboxed.sh
All your AI providers, machines, and agent harnesses in one place. Formerly sandboxed.sh (now the backend to the Orb clients).
Th0rgal/sandboxed.sh
Boss skill for parallel worker orchestration. Analyze, split, delegate by outcome, judge by acceptance criteria, integrate. Do not implement directly.
Th0rgal/sandboxed.sh
Worker skill for boss-spawned missions. Stay within scope, verify, and report blockers quickly.
affaan-m/everything-claude-code
Eval-driven development (EDD) framework for AI coding sessions — define capability and regression evals before coding, grade with code-based, model-based, rule, or human graders, and track pass@k and pass^k reliability.…
affaan-m/everything-claude-code
Map a described workflow to the right ECC command group with run-order and stop condition, or browse all command-group recipe families read live from the commands directory. Advisory only — never executes. Use when…
affaan-m/everything-claude-code
Design task-local harnesses, eval gates, and reusable skill extraction for Claude dynamic workflow mode and other adaptive agent harnesses. Use when building a task-local harness, adding eval gates, or extracting a…
Add this badge to your README
[](https://modelsatlas.com/skills/instavm/coderunner/)
<a href="https://modelsatlas.com/skills/instavm/coderunner/"><img src="https://modelsatlas.com/skills/instavm/coderunner/badge.svg" alt="ModelsAtlas: CAUTION B"></a>
Questions about coderunner
What does the coderunner skill do?
Is coderunner safe to install?
What can coderunner access on my machine?
Which assistants does coderunner work with?
How current is this page?
Sources
- github.com/instavm/coderunner — the audited source at commit 132da8183323 and every finding aboveObserved · 12 Nov 2026
Spotted something wrong? Report a correction · Methodology
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- What does the coderunner skill do?
- Is coderunner safe to install?
- What can coderunner access on my machine?