Skills model-discovery
📦
model-discovery
Content revision r1 High Risk 🔑 Env variables🌐 Network access📁 Filesystem access⚙️ External commands
Discover Current AI Models and Tiers
Model identifiers change and quickly make configurations stale. This skill queries supported providers, caches results, detects additions, and recommends capability tiers.
Supports: Claude Codex Code(CC)
Stop for confirmation before installing.
Review the plan and obtain explicit user consent before changing files.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "model-discovery" from https://skillstore.io/skills/consiliency-model-discovery.md and its manifest at https://skillstore.io/api/skills/consiliency-model-discovery/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.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.
Test it
Using "model-discovery". Show current Anthropic model tiers from the local configuration.
Expected outcome:
- Fast: Claude Haiku 4.5.
- Default: Claude Sonnet 4.5.
- Heavy: Claude Opus 4.5.
Using "model-discovery". Check whether any fetched OpenAI models need classification.
Expected outcome:
Two new OpenAI models were detected. One matches the fast pattern, while one needs manual classification.
Using "model-discovery". Explain why Ollama discovery returned no models.
Expected outcome:
The configured Ollama host returned an empty model list. Confirm the service is running and that models have been pulled.
Security Audit
High Riskv9 • 7/13/2026 Open versioned report
No malicious intent or prompt injection was found, and most alerts are documentation or language-token false positives. Confirmed risks include broad parent .env loading, credentials in curl arguments or URLs, an unvalidated Ollama host, and a predictable temporary path.
11
Files scanned
1,921
Lines analyzed
22
Review items
0
False positives ignored
Confirmed security concerns (2)
High
Environment file access
env_file = parent / ".env"
The loop probes every parent directory for a .env file, and the following load call imports the first match. This scope can capture unrelated project secrets.
Medium
Unvalidated Ollama Host Allows Arbitrary Requests
OLLAMA_HOST is used without scheme or host validation. Inherited configuration can direct curl to an arbitrary network host at the fixed API path.
The environment value is concatenated directly into the curl URL with no allowlist or URL validation. The request occurs whenever Ollama fetching is enabled.
Capability review items (22)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
High
Generic API/secret keys
-H "x-api-key: $ANTHROPIC_API_KEY" \
The documented shell command expands the API key into a curl header argument. Command arguments can expose the credential through process listings.
High
Generic API/secret keys
-H "x-api-key: $ANTHROPIC_API_KEY" \
The documented shell command expands the API key into a curl header argument. Command arguments can expose the credential through process listings.
High
Generic API/secret keys
-H "x-api-key: $ANTHROPIC_API_KEY" \
The documented shell command expands the API key into a curl header argument. Command arguments can expose the credential through process listings.
High
Generic API/secret keys
API key as query parameter: `?key=$GEMINI_API_KEY`
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$GEMINI_API_KEY"
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$GEMINI_API_KEY" | \
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$GEMINI_API_KEY" | \
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$GEMINI_API_KEY" | \
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$GEMINI_API_KEY&pageSize=100"
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$GEMINI_API_KEY&pageToken=NEXT_
The guidance places the Gemini API key in a URL query parameter. Query strings and expanded curl arguments can leak through histories, logs, and process listings.
High
Generic API/secret keys
-H "Authorization: Bearer $OPENAI_API_KEY"
The documented shell command expands the bearer token into a curl argument. Other local users may observe command arguments through process listings.
High
Generic API/secret keys
-H "Authorization: Bearer $OPENAI_API_KEY" | \
The documented shell command expands the bearer token into a curl argument. Other local users may observe command arguments through process listings.
High
Generic API/secret keys
-H "Authorization: Bearer $OPENAI_API_KEY" | \
The documented shell command expands the bearer token into a curl argument. Other local users may observe command arguments through process listings.
High
Generic API/secret keys
-H "Authorization: Bearer $OPENAI_API_KEY" | \
The documented shell command expands the bearer token into a curl argument. Other local users may observe command arguments through process listings.
High
Generic API/secret keys
"-H", f"x-api-key: {api_key}",
The Anthropic API key is interpolated into a curl command argument. Process listings can reveal the expanded header to other local users.
High
Generic API/secret keys
"-H", f"Authorization: Bearer {api_key}"
The OpenAI bearer token is interpolated into a curl command argument. Process listings can reveal the expanded header to other local users.
High
Generic API/secret keys
url = f"https://generativelanguage.googleapis.com/v1beta/models?key={api_key}"
The Gemini key is embedded in the request URL and then passed to curl. URLs can expose credentials through process listings, logs, and diagnostics.
Medium
Temp directory access
if ! uv run python scripts/check_new_models.py --json > /tmp/new_models.json 2>&1; then
The workflow redirects output to a predictable shared /tmp path. A pre-existing symlink can redirect the write and overwrite another user-accessible file.
Low
dotenv library
load_dotenv(env_file)
This call automatically imports values from a discovered .env file into the process. That implicit credential loading can use secrets without explicit per-run consent.
Low
dotenv library
load_dotenv()
This call automatically imports values from a discovered .env file into the process. That implicit credential loading can use secrets without explicit per-run consent.
Low
Python dotenv loader
load_dotenv(env_file)
This call automatically imports values from a discovered .env file into the process. That implicit credential loading can use secrets without explicit per-run consent.
Low
Python dotenv loader
load_dotenv()
This call automatically imports values from a discovered .env file into the process. That implicit credential loading can use secrets without explicit per-run consent.
Risk Factors
🔑 Env variables (46)
config/model_tiers.json:25 config/model_tiers.json:58 config/model_tiers.json:101 config/model_tiers.json:134 cookbook/anthropic-models.md:15 cookbook/anthropic-models.md:22 cookbook/anthropic-models.md:31 cookbook/anthropic-models.md:37 cookbook/anthropic-models.md:93 cookbook/gemini-models.md:13 cookbook/gemini-models.md:18 cookbook/gemini-models.md:25 cookbook/gemini-models.md:29 cookbook/gemini-models.md:33 cookbook/gemini-models.md:89 cookbook/gemini-models.md:92 cookbook/openai-models.md:15 cookbook/openai-models.md:21 cookbook/openai-models.md:29 cookbook/openai-models.md:34 cookbook/openai-models.md:39 cookbook/openai-models.md:103 scripts/fetch_models.py:45 scripts/fetch_models.py:73 scripts/fetch_models.py:115 scripts/fetch_models.py:164 scripts/fetch_models.py:23 scripts/fetch_models.py:30 scripts/fetch_models.py:34 scripts/fetch_models.py:36 scripts/fetch_models.py:23 scripts/fetch_models.py:30 scripts/fetch_models.py:34 scripts/fetch_models.py:73 scripts/fetch_models.py:74 scripts/fetch_models.py:75 scripts/fetch_models.py:81 scripts/fetch_models.py:115 scripts/fetch_models.py:116 scripts/fetch_models.py:117 scripts/fetch_models.py:123 scripts/fetch_models.py:163 scripts/fetch_models.py:164 scripts/fetch_models.py:165 scripts/fetch_models.py:166 scripts/fetch_models.py:169
🌐 Network access (31)
cookbook/anthropic-models.md:8 cookbook/anthropic-models.md:21 cookbook/anthropic-models.md:30 cookbook/anthropic-models.md:36 cookbook/gemini-models.md:8 cookbook/gemini-models.md:18 cookbook/gemini-models.md:25 cookbook/gemini-models.md:29 cookbook/gemini-models.md:33 cookbook/gemini-models.md:89 cookbook/gemini-models.md:92 cookbook/ollama-models.md:10 cookbook/ollama-models.md:16 cookbook/ollama-models.md:28 cookbook/ollama-models.md:35 cookbook/ollama-models.md:38 cookbook/ollama-models.md:42 cookbook/ollama-models.md:128 cookbook/ollama-models.md:131 cookbook/ollama-models.md:139 cookbook/ollama-models.md:139 cookbook/openai-models.md:8 cookbook/openai-models.md:20 cookbook/openai-models.md:28 cookbook/openai-models.md:33 cookbook/openai-models.md:38 scripts/fetch_models.py:45 scripts/fetch_models.py:80 scripts/fetch_models.py:122 scripts/fetch_models.py:169 SKILL.md:19
📁 Filesystem access (3)
⚙️ External commands (33)
scripts/fetch_models.py:78 scripts/fetch_models.py:120 scripts/fetch_models.py:170 scripts/fetch_models.py:206 SKILL.md:33 SKILL.md:45 SKILL.md:47 SKILL.md:53 SKILL.md:55 SKILL.md:56 SKILL.md:61 SKILL.md:123-141 SKILL.md:141-147 SKILL.md:147-148 SKILL.md:148-149 SKILL.md:149-155 SKILL.md:155 SKILL.md:156-157 SKILL.md:157 SKILL.md:158-164 SKILL.md:164-174 SKILL.md:174-178 SKILL.md:178-193 SKILL.md:193-199 SKILL.md:199-202 SKILL.md:202-203 SKILL.md:203-204 SKILL.md:204-205 SKILL.md:205-208 SKILL.md:208-216 SKILL.md:216 SKILL.md:217 SKILL.md:226-234
Audited by: codex 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.
Copy report link
https://skillstore.io/skills/consiliency-model-discovery/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/consiliency-model-discovery?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/consiliency-model-discovery?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/consiliency-model-discovery/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/consiliency-model-discovery.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
Consiliency. (2026). model-discovery security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/consiliency-model-discovery/audits/9BibTeX citation
@techreport{consiliency-consiliency-model-discovery-2026,
author = {Consiliency},
title = {model-discovery security audit report (audit version 9)},
institution = {Skillstore},
year = {2026},
number = {9},
url = {https://skillstore.io/skills/consiliency-model-discovery/audits/9},
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: "model-discovery security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
- name: "Consiliency"
date-released: "2026-07-13"
url: "https://skillstore.io/skills/consiliency-model-discovery/audits/9"
identifiers:
- type: other
value: "skillstore:consiliency-model-discovery:audit:9"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: Medium 45
Architecture
85
Maintainability
87
Content
70
Community
83
Spec Compliance
What You Can Build
Refresh application model identifiers
Check provider APIs before updating production model configuration or deployment settings.
Choose a task-appropriate model tier
Compare fast, default, and heavy recommendations for latency, cost, and capability needs.
Monitor newly available models
Detect unclassified provider models and review suggested tiers before adding them to the registry.
Try These Prompts
List cached models
Show the cached models for each provider. Include the fetch timestamp and any provider errors.
Refresh one provider
Refresh models for [provider]. Summarize available identifiers and note any authentication or connection errors.
Recommend model tiers
Review current models for [provider]. Recommend fast, default, and heavy options with concise reasons based on available metadata.
Audit and classify new models
Refresh all enabled providers, detect unknown models, and propose tiers. Do not update the registry until I approve each uncertain classification.
Best Practices
- Review cached timestamps before making network requests.
- Inspect provider errors before using returned model identifiers.
- Approve uncertain tier assignments manually before updating the known-model registry.
Avoid
- Do not hardcode provider model versions without checking current availability.
- Do not place API credentials in prompts, logs, or shared output.
- Do not auto-classify unfamiliar models when no naming pattern matches.
Frequently Asked Questions
Which providers are supported?
The skill supports Anthropic, OpenAI, Google Gemini, and Ollama.
Which credentials are required?
Anthropic, OpenAI, and Gemini require their documented API keys. Ollama requires no key by default.
Does the skill cache provider responses?
Yes. It stores model results locally and treats them as fresh for 24 hours.
How are models assigned to tiers?
Configurable name patterns infer fast, default, or heavy tiers. Unmatched models can be reviewed interactively.
Can it update classifications automatically?
Yes. Automatic mode saves inferred tiers, while dry-run mode reports changes without writing them.
Does it verify model quality or pricing?
No. It discovers model metadata and applies naming rules. It does not benchmark quality, latency, pricing, or availability guarantees.
Developer Details
Author
ConsiliencyLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Repository
https://github.com/Consiliency/treesitter-chunker/tree/main/.ai-dev-kit/skills/model-discoveryRef
635f69fb8d2f4e6330ba47a4e5a0fb239c04d110
Maintenance freshness
7/18/2026
Usage
6 downloads · 178 views
File structure